gm(1) General Commands Manual gm(1)
NAME
gm - command-line utility to create, edit, compare, convert, or
display images
SYNOPSIS
gm animate [ options ... ] file [ [ options ... ] file ... ]
gm batch [ options ... ] [ script ]
gm benchmark [ options ... ] subcommand
gm compare [ options ... ] reference-image [ options ... ] compare-im-
age [ options ... ]
gm composite [ options ... ] change-image base-image [ mask-image ]
output-image
gm conjure [ options ] script.msl [ [ options ] script.msl ]
gm convert [ [ options ... ] [ input-file ... ] [ options ... ] ] out-
put-file
gm display [ options ... ] file ... [ [options ... ]file ... ]
gm identify file [ file ... ]
gm import [ options ... ] file
gm mogrify [ options ... ] file ...
gm montage [ options ... ] file [ [ options ... ] file ... ] output-
file
gm time subcommand
gm version
DESCRIPTION
GraphicsMagick's gm provides a suite of utilities for creating, compar-
ing, converting, editing, and displaying images. All of the utilities
are provided as sub-commands of a single gm executable. The gm exe-
cutable returns the exit code 0 to indicate success, or 1 to indicate
failure:
animate displays an animation (e.g. a GIF file) on any workstation dis-
play running an X server.
batch executes an arbitary number of the utility commands (e.g. con-
vert) in the form of a simple linear batch script in order to improve
execution efficiency, and/or to allow use as a subordinate co-process
under the control of an arbitrary script or program.
benchmark executes one of the other utility commands (e.g. convert) for
a specified number of iterations, or execution time, and reports execu-
tion time and other profiling information such as CPU utilization.
Benchmark provides various operating modes including executing the com-
mand with a varying number of threads, and alternate reporting formats
such as comma-separated value (CSV).
compare compares two images and reports difference statistics according
to specified metrics and/or outputs an image with a visual representa-
tion of the differences. It may also be used to test if images are
similar within a particular range and specified metric, returning a
truth value to the executing environment.
composite composites images (blends or merges images together) to cre-
ate new images.
conjure interprets and executes scripts in the Magick Scripting Lan-
guage (MSL).
convert converts an input file using one image format to an output file
with the same or differing image format while applying an arbitrary
number of image transformations.
display is a machine architecture independent image processing and dis-
play facility. It can display an image on any workstation display run-
ning an X server.
identify describes the format and characteristics of one or more image
files. It will also report if an image is incomplete or corrupt.
import reads an image from any visible window on an X server and out-
puts it as an image file. You can capture a single window, the entire
screen, or any rectangular portion of the screen.
mogrify transforms an image or a sequence of images. These transforms
include image scaling, image rotation, color reduction, and others. The
transmogrified image overwrites the original image.
montage creates a composite by combining several separate images. The
images are tiled on the composite image with the name of the image op-
tionally appearing just below the individual tile.
time executes a subcommand and reports the user, system, and total exe-
cution time consumed.
version reports the GraphicsMagick release version, maximum sample-
depth, copyright notice, supported features, and the options used while
building the software.
The GraphicsMagick utilities recognize the following image formats:
Name Mode Description
o 3FR r-- Hasselblad Photo RAW
o 8BIM rw- Photoshop resource format
o 8BIMTEXT rw- Photoshop resource text format
o 8BIMWTEXT rw- Photoshop resource wide text format
o APP1 rw- Raw application information
o APP1JPEG rw- Raw JPEG binary data
o ART r-- PF1: 1st Publisher
o ARW r-- Sony Alpha DSLR RAW
o AVS rw+ AVS X image
o BIE rw- Joint Bi-level Image experts Group
interchange format
o BMP rw+ Microsoft Windows bitmap image
o BMP2 -w- Microsoft Windows bitmap image v2
o BMP3 -w- Microsoft Windows bitmap image v3
o CACHE --- Magick Persistent Cache image format
o CALS rw- Continuous Acquisition and Life-cycle
Support Type 1 image
o CAPTION r-- Caption (requires separate size info)
o CIN rw- Kodak Cineon Format
o CMYK rw- Raw cyan, magenta, yellow, and black
samples (8 or 16 bits, depending on
the image depth)
o CMYKA rw- Raw cyan, magenta, yellow, black, and
matte samples (8 or 16 bits, depending
on the image depth)
o CR2 r-- Canon Photo RAW
o CRW r-- Canon Photo RAW
o CUR r-- Microsoft Cursor Icon
o CUT r-- DR Halo
o DCM r-- Digital Imaging and Communications in
Medicine image
o DCR r-- Kodak Photo RAW
o DCX rw+ ZSoft IBM PC multi-page Paintbrush
o DNG r-- Adobe Digital Negative
o DPS r-- Display PostScript Interpreter
o DPX rw- Digital Moving Picture Exchange
o EPDF rw- Encapsulated Portable Document Format
o EPI rw- Adobe Encapsulated PostScript
Interchange format
o EPS rw- Adobe Encapsulated PostScript
o EPS2 -w- Adobe Level II Encapsulated PostScript
o EPS3 -w- Adobe Level III Encapsulated PostScript
o EPSF rw- Adobe Encapsulated PostScript
o EPSI rw- Adobe Encapsulated PostScript
Interchange format
o EPT rw- Adobe Encapsulated PostScript with MS-DOS
TIFF preview
o EPT2 rw- Adobe Level II Encapsulated PostScript
with MS-DOS TIFF preview
o EPT3 rw- Adobe Level III Encapsulated PostScript
with MS-DOS TIFF preview
o EXIF rw- Exif digital camera binary data
o FAX rw+ Group 3 FAX (Not TIFF Group3 FAX!)
o FITS rw- Flexible Image Transport System
o FRACTAL r-- Plasma fractal image
o FPX rw- FlashPix Format
o GIF rw+ CompuServe graphics interchange format
o GIF87 rw- CompuServe graphics interchange format
(version 87a)
o GRADIENT r-- Gradual passing from one shade to
another
o GRAY rw+ Raw gray samples (8/16/32 bits,
depending on the image depth)
o HISTOGRAM -w- Histogram of the image
o HRZ r-- HRZ: Slow scan TV
o HTML -w- Hypertext Markup Language and a
client-side image map
o ICB rw+ Truevision Targa image
o ICC rw- ICC Color Profile
o ICM rw- ICC Color Profile
o ICO r-- Microsoft icon
o ICON r-- Microsoft icon
o IDENTITY r-- Hald CLUT identity image
o IMAGE r-- GraphicsMagick Embedded Image
o INFO -w+ Image descriptive information and
statistics
o IPTC rw- IPTC Newsphoto
o IPTCTEXT rw- IPTC Newsphoto text format
o IPTCWTEXT rw- IPTC Newsphoto wide text format
o JBG rw+ Joint Bi-level Image experts Group
interchange format
o JBIG rw+ Joint Bi-level Image experts Group
interchange format
o JNG rw- JPEG Network Graphics
o JP2 rw- JPEG-2000 JP2 File Format Syntax
o JPC rw- JPEG-2000 Code Stream Syntax
o JPEG rw- Joint Photographic Experts Group
JFIF format
o JPG rw- Joint Photographic Experts Group
JFIF format
o K25 r-- Kodak Photo RAW
o KDC r-- Kodak Photo RAW
o LABEL r-- Text image format
o M2V rw+ MPEG-2 Video Stream
o MAP rw- Colormap intensities and indices
o MAT r-- MATLAB image format
o MATTE -w+ MATTE format
o MIFF rw+ Magick Image File Format
o MNG rw+ Multiple-image Network Graphics
o MONO rw- Bi-level bitmap in least-significant-
-byte-first order
o MPC rw+ Magick Persistent Cache image format
o MPEG rw+ MPEG-1 Video Stream
o MPG rw+ MPEG-1 Video Stream
o MRW r-- Minolta Photo Raw
o MSL r-- Magick Scripting Language
o MTV rw+ MTV Raytracing image format
o MVG rw- Magick Vector Graphics
o NEF r-- Nikon Electronic Format
o NULL r-- Constant image of uniform color
o OTB rw- On-the-air bitmap
o P7 rw+ Xv thumbnail format
o PAL rw- 16bit/pixel interleaved YUV
o PALM rw- Palm Pixmap
o PBM rw+ Portable bitmap format (black and white)
o PCD rw- Photo CD
o PCDS rw- Photo CD
o PCL -w- Page Control Language
o PCT rw- Apple Macintosh QuickDraw/PICT
o PCX rw- ZSoft IBM PC Paintbrush
o PDB rw+ Palm Database ImageViewer Format
o PDF rw+ Portable Document Format
o PEF r-- Pentax Electronic File
o PFA r-- TrueType font
o PFB r-- TrueType font
o PGM rw+ Portable graymap format (gray scale)
o PGX r-- JPEG-2000 VM Format
o PICON rw- Personal Icon
o PICT rw- Apple Macintosh QuickDraw/PICT
o PIX r-- Alias/Wavefront RLE image format
o PLASMA r-- Plasma fractal image
o PNG rw- Portable Network Graphics
o PNG24 rw- Portable Network Graphics, 24 bit RGB
opaque only
o PNG32 rw- Portable Network Graphics, 32 bit RGBA
semitransparency OK
o PNG8 rw- Portable Network Graphics, 8-bit
indexed, binary transparency only
o PNM rw+ Portable anymap
o PPM rw+ Portable pixmap format (color)
o PREVIEW -w- Show a preview an image enhancement,
effect, or f/x
o PS rw+ Adobe PostScript
o PS2 -w+ Adobe Level II PostScript
o PS3 -w+ Adobe Level III PostScript
o PSD rw- Adobe Photoshop bitmap
o PTIF rw- Pyramid encoded TIFF
o PWP r-- Seattle Film Works
o RAF r-- Fuji Photo RAW
o RAS rw+ SUN Rasterfile
o RGB rw+ Raw red, green, and blue samples
o RGBA rw+ Raw red, green, blue, and matte samples
o RLA r-- Alias/Wavefront image
o RLE r-- Utah Run length encoded image
o SCT r-- Scitex HandShake
o SFW r-- Seattle Film Works
o SGI rw+ Irix RGB image
o SHTML -w- Hypertext Markup Language and a
client-side image map
o STEGANO r-- Steganographic image
o SUN rw+ SUN Rasterfile
o SVG rw+ Scalable Vector Gaphics
o TEXT rw+ Raw text
o TGA rw+ Truevision Targa image
o TIFF rw+ Tagged Image File Format
o TILE r-- Tile image with a texture
o TIM r-- PSX TIM
o TOPOL r-- TOPOL X Image
o TTF r-- TrueType font
o TXT rw+ Raw text
o UIL -w- X-Motif UIL table
o UYVY rw- 16bit/pixel interleaved YUV
o VDA rw+ Truevision Targa image
o VICAR rw- VICAR rasterfile format
o VID rw+ Visual Image Directory
o VIFF rw+ Khoros Visualization image
o VST rw+ Truevision Targa image
o WBMP rw- Wireless Bitmap (level 0) image
o WMF r-- Windows Metafile
o WPG r-- Word Perfect Graphics
o X rw- X Image
o X3F r-- Foveon X3 (Sigma/Polaroid) RAW
o XBM rw- X Windows system bitmap (black
and white)
o XC r-- Constant image uniform color
o XCF r-- GIMP image
o XMP rw- Adobe XML metadata
o XPM rw- X Windows system pixmap (color)
o XV rw+ Khoros Visualization image
o XWD rw- X Windows system window dump (color)
o YUV rw- CCIR 601 4:1:1 or 4:2:2 (8-bit only)
Modes:
r Read
w Write
+ Multi-image
Support for some of these formats require additional programs or li-
braries. See README in the source package for where to find optional
additional software.
Note, a format delineated with + means that if more than one image is
specified, frames are combined into a single multi-image file. Use +ad-
join if you want a single image produced for each frame.
Your installation might not support all of the formats in the list. To
get an accurate listing of the formats supported by your particular
configuration, run "gm convert -list format".
Raw images are expected to have one byte per pixel unless gm is com-
piled in 16-bit quantum mode or in 32-bit quantum mode. Here, the raw
data is expected to be stored two or four bytes per pixel, respec-
tively, in most-significant-byte-first order. For example, you can
tell if gm was compiled in 16-bit mode by typing "gm version" without
any options, and looking for "Q:16" in the first line of output.
FILES AND FORMATS
By default, the image format is determined by its magic number, i.e.,
the first few bytes of the file. To specify a particular image format,
precede the filename with an image format name and a colon (i.e.ps:im-
age) or specify the image type as the filename suffix (i.e.image.ps).
The magic number takes precedence over the filename suffix and the pre-
fix takes precedence over the magic number and the suffix in input
files. When a file is read, its magic number is stored in the "im-
age->magick" string. In output files, the prefix takes precedence over
the filename suffix, and the filename suffix takes precedence over the
"image->magick" string.
To read the "built-in" formats (GRANITE, H, LOGO, NETSCAPE, PLASMA, and
ROSE) use a prefix (including the colon) without a filename or suffix.
To read the XC format, follow the colon with a color specification. To
read the CAPTION format, follow the colon with a text string or with a
filename prefixed with the at symbol (@).
When you specify X as your image type, the filename has special mean-
ing. It specifies an X window by id, name, or root. If no filename is
specified, the window is selected by clicking the mouse in the desired
window.
Specify input_file as - for standard input, output_file as - for stan-
dard output. If input_file has the extension .Z or .gz, the file is
uncompressed with uncompress or gunzip respectively. If output_file
has the extension .Z or .gz, the file is compressed using with compress
or gzip respectively.
Use an optional index enclosed in brackets after an input file name to
specify a desired subimage of a multi-resolution image format like
Photo CD (e.g. "img0001.pcd[4]") or a range for MPEG images (e.g.
"video.mpg[50-75]"). A subimage specification can be disjoint (e.g.
"image.tiff[2,7,4]"). For raw images, specify a subimage with a geome-
try (e.g. -size 640x512 "image.rgb[320x256+50+50]"). Surround the im-
age name with quotation marks to prevent your shell from interpreting
the square brackets. Single images are written with the filename you
specify. However, multi-part images (e.g., a multi-page PostScript doc-
ument with +adjoin specified) may be written with the scene number in-
cluded as part of the filename. In order to include the scene number in
the filename, it is necessary to include a printf-style %d format spec-
ification in the file name and use the +adjoin option. For example,
image%02d.miff
writes files image00.miff, image01.miff, etc. Only a single specifica-
tion is allowed within an output filename. If more than one specifica-
tion is present, it will be ignored. It is best to embed the scene num-
ber in the base part of the file name, not in the extension, because
the extension will not be a recognizeable image type.
When running a commandline utility, you can prepend an at sign @ to a
filename to read a list of image filenames from that file. This is con-
venient in the event you have too many image filenames to fit on the
command line.
OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect for the set of images that follows,
until the set is terminated by the appearance of any option or -noop.
Some options only affect the decoding of images and others only the en-
coding. The latter can appear after the final group of input images.
This is a combined list of the command-line options used by the Graph-
icsMagick utilities (animate, compare, composite, convert, display,
identify, import, mogrify and montage).
In this document, angle brackets ("<>") enclose variables and curly
brackets ("{}") enclose optional parameters. For example, "-fuzz <dis-
tance>{%}" means you can use the option "-fuzz 10" or "-fuzz 2%".
-adjoin
join images into a single multi-image file
By default, all images of an image sequence are stored in the
same file. However, some formats (e.g. JPEG) do not support
storing more than one image per file and only the first frame in
an image sequence will be saved unless the result is saved to
separate files. Use +adjoin to force saving multiple frames to
multiple numbered files. If +adjoin is used, then the output
filename must include a printf style formatting specification
for the numeric part of the filename. For example,
image%02d.miff
-affine <matrix>
drawing transform matrix
This option provides a transform matrix {sx,rx,ry,sy,tx,ty} for
use by subsequent -draw or -transform options.
-antialias
remove pixel aliasing
By default antialiasing algorithms are used when drawing objects
(e.g. lines) or rendering vector formats (e.g. WMF and Post-
script). Use +antialias to disable use of antialiasing algo-
rithms. Reasons to disable antialiasing include avoiding in-
creasing colors in the image, or improving rendering speed.
-append
append a set of images
This option creates a single image where the images in the orig-
inal set are stacked top-to-bottom. If they are not of the same
width, any narrow images will be expanded to fit using the back-
ground color. Use +append to stack images left-to-right. The
set of images is terminated by the appearance of any option. If
the -append option appears after all of the input images, all
images are appended.
-asc-cdl <spec>
apply ASC CDL color transform
Applies ("bakes in") the ASC CDL, which is a format for the ex-
change of basic primary color grading information between equip-
ment and software from different manufacturers. The format de-
fines the math for three functions: slope, offset and power.
Each function uses a number for the red, green, and blue color
channels for a total of nine numbers comprising a single color
decision. The tenth number (optional) is for chromiance (satura-
tion) as specified by ASC CDL 1.2.
The argument string is comma delimited and is in the following
form (but without invervening spaces or line breaks)
redslope,redoffset,redpower:
greenslope,greenoffset,greenpower:
blueslope,blueoffset,bluepower:
saturation
with the unity (no change) specification being:
"1.0,0.0,1.0:1.0,0.0,1.0:1.0,0.0,1.0:1.0"
-authenticate <string>
decrypt image with this password
Use this option to supply a password for decrypting an image or
an image sequence, if it is being read from a format such as PDF
that supports encryption. Encrypting images being written is
not supported.
-auto-orient
orient (rotate) image so it is upright
Adjusts the image orienation so that it is suitable for viewing.
Uses the orientation tag obtained from the image file or as sup-
plied by the -orient option.
-average
average a set of images
The set of images is terminated by the appearance of any option.
If the -average option appears after all of the input images,
all images are averaged.
-backdrop
display the image centered on a backdrop.
This backdrop covers the entire workstation screen and is useful
for hiding other X window activity while viewing the image. The
color of the backdrop is specified as the foreground color (X11
default is black). Refer to "X Resources", below, for details.
-background <color>
the background color
The color is specified using the format described under the
-fill option.
-black-threshold red[,green][,blue][,opacity]
pixels below the threshold become black
Use -black-threshold to set pixels with values below the speci-
fied threshold to minimum value (black). If only one value is
supplied, or the red, green, and blue values are identical, then
intensity thresholding is used. If the color threshold values
are not identical then channel-based thresholding is used, and
color distortion will occur. Specify a negative value (e.g. -1)
if you want a channel to be ignored but you do want to threshold
a channel later in the list. If a percent (%) symbol is ap-
pended, then the values are treated as a percentage of maximum
range.
-blue-primary <x>,<y>
blue chromaticity primary point
-blur <radius>{x<sigma>}
blur the image with a Gaussian operator
Blur with the given radius and standard deviation (sigma).
-border <width>x<height>
surround the image with a border of color
See -geometry for details about the geometry specification.
-bordercolor <color>
the border color
The color is specified using the format described under the
-fill option.
-borderwidth <geometry>
the border width
-box <color>
set the color of the annotation bounding box
The color is specified using the format described under the
-fill option.
See -draw for further details.
-channel <type>
the type of channel
Choose from: Red, Green, Blue, Opacity, Matte, Cyan, Magenta,
Yellow, Black, or Gray.
Use this option to extract a particular channel from the image.
Opacity, for example, is useful for extracting the opacity val-
ues from an image.
-charcoal <factor>
simulate a charcoal drawing
-chop <width>x<height>{+-}<x>{+-}<y>{%}
remove pixels from the interior of an image
Width and height give the number of columns and rows to remove,
and x and y are offsets that give the location of the leftmost
column and topmost row to remove.
The x offset normally specifies the leftmost column to remove.
If the -gravity option is present with NorthEast, East, or
SouthEast gravity, it gives the distance leftward from the right
edge of the image to the rightmost column to remove. Similarly,
the y offset normally specifies the topmost row to remove, but
if the -gravity option is present with SouthWest, South, or
SouthEast gravity, it specifies the distance upward from the
bottom edge of the image to the bottom row to remove.
The -chop option removes entire rows and columns, and moves the
remaining corner blocks leftward and upward to close the gaps.
-clip apply the clipping path, if one is present
If a clipping path is present, it will be applied to subsequent
operations.
For example, if you type the following command:
gm convert -clip -negate cockatoo.tif negated.tif
only the pixels within the clipping path are negated.
The -clip feature requires the XML library. If the XML library
is not present, the option is ignored.
-coalesce
merge a sequence of images
Each image N in the sequence after Image 0 is replaced with the
image created by flattening images 0 through N.
The set of images is terminated by the appearance of any option.
If the -coalesce option appears after all of the input images,
all images are coalesced.
-colorize <value>
colorize the image with the pen color
Specify the amount of colorization as a percentage. You can ap-
ply separate colorization values to the red, green, and blue
channels of the image with a colorization value list delimited
with slashes (e.g. 0/0/50).
The -colorize option may be used in conjunction with -modulate
to produce a nice sepia toned image like:
gm convert input.ppm -modulate 115,0,100 \
-colorize 7,21,50 output.ppm.
-colormap <type>
define the colormap type
Choose between shared or private.
This option only applies when the default X server visual is
PseudoColor or GRAYScale. Refer to -visual for more details. By
default, a shared colormap is allocated. The image shares colors
with other X clients. Some image colors could be approximated,
therefore your image may look very different than intended.
Choose Private and the image colors appear exactly as they are
defined. However, other clients may go technicolor when the im-
age colormap is installed.
-colors <value>
preferred number of colors in the image
The actual number of colors in the image may be less than your
request, but never more. Note, this is a color reduction option.
Images with less unique colors than specified with this option
will have any duplicate or unused colors removed. The ordering
of an existing color palette may be altered. When converting an
image from color to grayscale, convert the image to the gray
colorspace before reducing the number of colors since doing so
is most efficient. Refer to <a href="quantize.html">quantize for
more details.
Note, options -dither, -colorspace, and -treedepth affect the
color reduction algorithm.
-colorspace <value>
the type of colorspace
Choices are: CineonLog, CMYK, GRAY, HSL, HWB, OHTA, RGB,
Rec601Luma, Rec709Luma, Rec601YCbCr, Rec709YCbCr, Transparent,
XYZ, YCbCr, YIQ, YPbPr, or YUV.
Color reduction, by default, takes place in the RGB color space.
Empirical evidence suggests that distances in color spaces such
as YUV or YIQ correspond to perceptual color differences more
closely than do distances in RGB space. These color spaces may
give better results when color reducing an image. Refer to
quantize for more details. Two gray colorspaces are supported.
The Rec601Luma space is based on the recommendations for legacy
NTSC television (ITU-R BT.601-5). The Rec709Luma space is based
on the recommendations for HDTV (Rec. ITU-R BT.709-5) and is
suitable for use with computer graphics, and for contemporary
CRT displays. The GRAY colorspace currently selects the
Rec601Luma colorspace by default for backwards compatibly rea-
sons. This default may be re-considered in the future.
Two YCbCr colorspaces are supported. The Rec601YCbCr space is
based on the recommendations for legacy NTSC television (ITU-R
BT.601-5). The Rec709CbCr space is based on the recommendations
for HDTV (Rec. ITU-R BT.709-5) and is suitable for suitable for
use with computer graphics, and for contemporary CRT displays.
The YCbCr colorspace specification is equivalent toRec601YCbCr.
The Transparent color space behaves uniquely in that it pre-
serves the matte channel of the image if it exists.
The -colors or -monochrome option, or saving to a file format
which requires color reduction, is required for this option to
take effect.
-comment <string>
annotate an image with a comment
Use this option to assign a specific comment to the image, when
writing to an image format that supports comments. You can in-
clude the image filename, type, width, height, or other image
attribute by embedding special format characters listed under
the -format option. The comment is not drawn on the image, but
is embedded in the image datastream via a "Comment" tag or simi-
lar mechanism. If you want the comment to be visible on the im-
age itself, use the -draw option instead.
For example,
-comment "%m:%f %wx%h"
produces an image comment of MIFF:bird.miff 512x480 for an image
titled bird.miff and whose width is 512 and height is 480.
If the first character of string is @, the image comment is read
from a file titled by the remaining characters in the string.
Please note that if the string comes from an untrusted source
that it should be sanitized before use since otherwise the con-
tent of an arbitrary readable file could be incorporated in a
comment in the output file (a security risk).
If the -comment option appears multiple times, only the last
comment is stored.
In PNG images, the comment is stored in a tEXt or zTXt chunk
with the keyword "comment".
-compose <operator>
the type of image composition
The description of composition uses abstract terminology in or-
der to allow the the description to be more clear, while avoid-
ing constant values which are specific to a particular build
configuration. Each image pixel is represented by red, green,
and blue levels (which are equal for a gray pixel). MaxRGB is
the maximum integral value which may be stored in the red,
green, or blue channels of the image. Each image pixel may also
optionally (if the image matte channel is enabled) have an asso-
ciated level of opacity (ranging from opaque to transparent),
which may be used to determine the influence of the pixel color
when compositing the pixel with another image pixel. If the im-
age matte channel is disabled, then all pixels in the image are
treated as opaque. The color of an opaque pixel is fully visi-
ble while the color of a transparent pixel color is entirely ab-
sent (pixel color is ignored).
By definition, raster images have a rectangular shape. All image
rows are of equal length, and all image columns have the same
number of rows. By treating the opacity channel as a visual
"mask" the rectangular image may be given a "shape" by treating
the opacity channel as a cookie-cutter for the image. Pixels
within the shape are opaque, while pixels outside the shape are
transparent. Pixels on the boundary of the shape may be between
opaque and transparent in order to provide antialiasing (vis-
ually smooth edges). The description of the composition opera-
tors use this concept of image "shape" in order to make the de-
scription of the operators easier to understand. While it is
convenient to describe the operators in terms of "shapes" they
are by no means limited to mask-style operations since they are
based on continuous floating-point mathematics rather than sim-
ple boolean operations.
By default, the Over composite operator is used. The following
composite operators are available:
Over
In
Out
Atop
Xor
Plus
Minus
Add
Subtract
Difference
Divide
Multiply
Bumpmap
Copy
CopyRed
CopyGreen
CopyBlue
CopyOpacity
CopyCyan
CopyMagenta
CopyYellow
CopyBlack
The behavior of each operator is described below.
Over
The result will be the union of the two image shapes, with
opaque areas of change-image obscuring base-image in the
region of overlap.
In
The result is simply change-image cut by the shape of
base-image. None of the image data of base-image will be
in the result.
Out
The resulting image is change-image with the shape of
base-image cut out.
Atop
The result is the same shape as base-image, with change-
image obscuring base-image where the image shapes overlap.
Note this differs from over because the portion of change-
image outside base-image's shape does not appear in the
result.
Xor
The result is the image data from both change-image and
base-image that is outside the overlap region. The overlap
region will be blank.
Plus
The result is just the sum of the image data. Output val-
ues are cropped to MaxRGB (no overflow). This operation is
independent of the matte channels.
Minus
The result of change-image - base-image, with underflow
cropped to zero. The matte channel is ignored (set to
opaque, full coverage).
Add
The result of change-image + base-image, with overflow
wrapping around (mod MaxRGB+1).
Subtract
The result of change-image - base-image, with underflow
wrapping around (mod MaxRGB+1). The add and subtract oper-
ators can be used to perform reversible transformations.
Difference
The result of abs(change-image - base-image). This is use-
ful for comparing two very similar images.
Divide
The result of change-image / base-image. This is useful
for improving the readability of text on unevenly illumi-
nated photos (by using a gaussian blurred copy of change-
image as base-image).
Multiply
The result of change-image * base-image. This is useful
for the creation of drop-shadows.
Bumpmap
The result base-image shaded by change-image.
Copy
The resulting image is base-image replaced with change-im-
age. Here the matte information is ignored.
CopyRed
The resulting image is the red channel in base-image re-
placed with the red channel in change-image. The other
channels are copied untouched.
CopyGreen
The resulting image is the green channel in base-image re-
placed with the green channel in change-image. The other
channels are copied untouched.
CopyBlue
The resulting image is the blue channel in base-image re-
placed with the blue channel in change-image. The other
channels are copied untouched.
CopyOpacity
The resulting image is the opacity channel in base-image
replaced with the opacity channel in change-image. The
other channels are copied untouched.
CopyCyan
The resulting image is the cyan channel in base-image re-
placed with the cyan channel in change-image. The other
channels are copied untouched. Use of this operator re-
quires that base-image be in CMYK(A) colorspace.
CopyMagenta
The resulting image is the magenta channel in base-image
replaced with the magenta channel in change-image. The
other channels are copied untouched. Use of this operator
requires that base-image be in CMYK(A) colorspace.
CopyYellow
The resulting image is the yellow channel in base-image
replaced with the yellow channel in change-image. The
other channels are copied untouched. Use of this operator
requires that base-image be in CMYK(A) colorspace.
CopyBlack
The resulting image is the black channel in base-image re-
placed with the black channel in change-image. The other
channels are copied untouched. Use of this operator re-
quires that base-image be in CMYK(A) colorspace. If
change-image is not in CMYK space, then the change-image
pixel intensities are used.
-compress <type>
the type of image compression
Choices are: None, BZip, Fax, Group3, Group4, JPEG, Lossless,
LZW, RLE, Zip, LZMA, JPEG2000, JPEG2000, JBIG, JBIG2, WebP, or
ZSTD.
Specify +compress to store the binary image in an uncompressed
format. The default is the compression type of the specified
image file.
"Lossless" refers to lossless JPEG, which is only available if
the JPEG library has been patched to support it. Use of lossless
JPEG is generally not recommended.
Use the -quality option to set the compression level to be used
by the JPEG, JPEG-2000, PNG, MIFF, MPEG, and TIFF encoders. Use
the -sampling-factor option to set the sampling factor to be
used by the DPX, JPEG, MPEG, and YUV encoders for downsampling
the chroma channels.
-contrast
enhance or reduce the image contrast
This option enhances the intensity differences between the
lighter and darker elements of the image. Use -contrast to en-
hance the image or +contrast to reduce the image contrast.
For a more pronounced effect you can repeat the option:
gm convert rose: -contrast -contrast rose_c2.png
-convolve <kernel>
convolve image with the specified convolution kernel
The kernel is specified as a comma-separated list of floating
point values, ordered left-to right, starting with the top row.
The order of the kernel is determined by the square root of the
number of entries. Presently only square kernels are supported.
-create-directories
create output directory if required
Use this option with -output-directory if the input paths con-
tain subdirectories and it is desired to create similar subdi-
rectories in the output directory. Without this option, mogrify
will fail if the required output directory does not exist.
-crop <width>x<height>{+-}<x>{+-}<y>{%}
preferred size and location of the cropped image
See -geometry for details about the geometry specification.
The width and height give the size of the image that remains af-
ter cropping, and x and y are offsets that give the location of
the top left corner of the cropped image with respect to the
original image. To specify the amount to be removed, use -shave
instead.
If the x and y offsets are present, a single image is generated,
consisting of the pixels from the cropping region. The offsets
specify the location of the upper left corner of the cropping
region measured downward and rightward with respect to the upper
left corner of the image. If the -gravity option is present
with NorthEast, East, or SouthEast gravity, it gives the dis-
tance leftward from the right edge of the image to the right
edge of the cropping region. Similarly, if the -gravity option
is present with SouthWest, South, or SouthEast gravity, the dis-
tance is measured upward between the bottom edges.
If the x and y offsets are omitted, a set of tiles of the speci-
fied geometry, covering the entire input image, is generated.
The rightmost tiles and the bottom tiles are smaller if the
specified geometry extends beyond the dimensions of the input
image.
-cycle <amount>
displace image colormap by amount
Amount defines the number of positions each colormap entry is-
shifted.
-debug <events>
enable debug printout
The events parameter specifies which events are to be logged.
It can be either None, All, or a comma-separated list consisting
of one or more of the following domains: Annotate, Blob, Cache,
Coder, Configure, Deprecate, Error, Exception, FatalError, In-
formation, Locale, Option, Render, Resource, TemporaryFile,
Transform, User. Warning, or X11, For example, to log cache and
blob events, use
gm convert -debug "Cache,Blob" rose: rose.png
The "User" domain is normally empty, but developers can log
"User" events in their private copy of GraphicsMagick.
Use the -log option to specify the format for debugging output.
Use +debug to turn off all logging.
An alternative to using -debug is to use the MAGICK_DEBUG envi-
ronment variable. The allowed values for the MAGICK_DEBUG envi-
ronment variable are the same as for the -debug option.
-deconstruct
break down an image sequence into constituent parts
This option compares each image with the next in a sequence and
returns the maximum bounding region of any pixel differences it
discovers. This method can undo a coalesced sequence returned
by the -coalesce option, and is useful for removing redundant
information from a GIF or MNG animation.
The sequence of images is terminated by the appearance of any
option. If the -deconstruct option appears after all of the in-
put images, all images are deconstructed.
-define <key>{=<value>},...
add coder/decoder specific options This option creates one or
more definitions for coders and decoders to use while reading
and writing image data. Definitions may be passed to coders and
decoders to control options that are specific to certain image
formats. If value is missing for a definition, an empty-valued
definition of a flag will be created with that name. This is
used to control on/off options. Use +define <key>,... to remove
definitions previously created. Use +define "*" to remove all
existing definitions.
The following definitions may be created:
cineon:colorspace={rgb|cineonlog}
Use the cineon:colorspace option when reading a Cineon
file to specify the colorspace the Cineon file uses. This
overrides the colorspace type implied by the DPX header
(if any).
dpx:bits-per-sample=<value>
If the dpx:bits-per-sample key is defined, GraphicsMagick
will write DPX images with the specified bits per sample,
overriding any existing depth value. If this option is not
specified, then the value is based on the existing image
depth value from the original image file. The DPX standard
supports bits per sample values of 1, 8, 10, 12, and 16.
Many DPX readers demand a sample size of 10 bits with type
A padding (see below).
dpx:colorspace={rgb|cineonlog}
Use the dpx:colorspace option when reading a DPX file to
specify the colorspace the DPX file uses. This overrides
the colorspace type implied by the DPX header (if any).
dpx:packing-method={packed|a|b|lsbpad|msbpad}
DPX samples are output within 32-bit words. They may be
tightly packed end-to-end within the words ("packed"),
padded with null bits to the right of the sample ("a" or
"lsbpad"), or padded with null bits to the left of the
sample ("b" or "msbpad"). This option only has an effect
for sample sizes of 10 or 12 bits. If samples are not
packed, the DPX standard recommends type A padding. Many
DPX readers demand a sample size of 10 bits with type A
padding.
dpx:pixel-endian={lsb|msb}
Allows the user to specify the endian order of the pixels
when reading or writing the DPX files. Sometimes this is
useful if the file is (or must be) written incorrectly so
that the file header and the pixels use different endian-
ness.
dpx:swap-samples={true|false}
dpx:swap-samples-read={true|false}
dpx:swap-samples-write={true|false}
GraphicsMagick strives to adhere to the DPX standard but
certain aspects of the standard can be quite confusing. As
a result, some 10-bit DPX files have Red and Blue inter-
changed, or Cb and Cr interchanged due to an different in-
terpretation of the standard, or getting the wires
crossed. The swap-samples option may be supplied when
reading or writing in order to read or write using the
necessary sample order. Use swap-samples-read when swap-
ping should only occur in the reader, or swap-samples-
write when swapping should only occur in the writer.
gradient:direction={South|North|West|East|NorthWest|North-
East|SouthWest|SouthEast}
By default, the gradient coder produces a gradient from
top to bottom ("South"). Since GraphicsMagick 1.3.35, the
gradient direction may be specified to produce gradient
vectors according to a gravity-like specification. The
arguments are South (Top to Bottom), North (Bottom to
Top), West (Right to Left), East (Left to Right), North-
West (Bottom-Right to Top-Left), NorthEast (Bottom-Left to
Top-Right), SouthWest (Top-Right Bottom-Left), and South-
East (Top-Left to Bottom-Right).
jp2:rate=<value>
Specify the compression factor to use while writing
JPEG-2000 files. The compression factor is the reciprocal
of the compression ratio. The valid range is 0.0 to 1.0,
with 1.0 indicating lossless compression. If defined, this
value overrides the -quality setting. The default quality
setting of 75 results in a rate value of 0.06641.
jpeg:arithmetic={true|false}
Enables or disables arithmetic encoding if the JPEG li-
brary supports it (default disabled). When this is en-
abled, optimize-coding is necessarily disabled.
jpeg:block-smoothing={true|false}
Enables or disables block smoothing when reading a JPEG
file (default enabled).
jpeg:dct-method=<value>
Selects the IJG JPEG library DCT implementation to use.
The encoding implementations vary in speed and encoding
error. The available choices for value are islow, ifast,
float, default and fastest. Note that fastest might not
necessarily be fastest on your CPU, depending on the
choices made when the JPEG library was built and how your
CPU behaves.
jpeg:fancy-upsampling={true|false}
Enables or disables fancy upsampling when reading a JPEG
file (default enabled).
jpeg:max-scan-number=<value>
Specifies an integer value for the maximum number of pro-
gressive scans allowed in a JPEG file. The default maxi-
mum is 100 scans. This limit is imposed due to a weakness
in the JPEG standard which allows small JPEG files to take
many minutes or hours to be read.
jpeg:max-warnings=<value>
Specifies an integer value for how many warnings are al-
lowed for any given error type before being promoted to a
hard error. JPEG files producing excessive warnings indi-
cate a problem with the file.
jpeg:optimize-coding={true|false}
Selects if huffman encoding should be used. Huffman encod-
ing is enabled by default, but may be disabled for very
large images since it encoding requires that the entire
image be buffered in memory. Huffman encoding produces
smaller JPEG files at the expense of added compression
time and memory consumption.
jpeg:preserve-settings
If the jpeg:preserve-settings flag is defined, the JPEG
encoder will use the same "quality" and "sampling-factor"
settings that were found in the input file, if the input
was in JPEG format. These settings are also preserved if
the input is a JPEG file and the output is a JNG file. If
the colorspace of the output file differs from that of the
input file, the quality setting is preserved but the sam-
pling-factors are not.
pcl:fit-to-page
If the pcl:fit-to-page flag is defined, then the printer
is requested to scale the image to fit the page size
(width and/or height).
png:chunk-malloc-max=<value>
png:chunk-malloc-max specifies the maximum chunk size that
libpng will be allowed to read. Libpng's default is nor-
mally 8,000,000 bytes. Very rarely, a valid PNG file may
be encountered where the error is reported "chunk data is
too large". In this case, the limit may be increased us-
ing this option. Take care when increasing this limit
since an excessively large limit could allow untrusted
files to use excessive memory.
mng:maximum-loops=<value>
mng:maximum-loops specifies the maximum number of loops
allowed to be specified by a MNG LOOP chunk. Without an
imposed limit, a MNG file could request up to 2147483647
loops, which could run for a very long time. The current
default limit is 512 loops.
pdf:use-cropbox={true|false}
If the pdf:use-cropbox flag is set to true, then Ghost-
script is requested to apply the PDF crop box.
pdf:stop-on-error={true|false}
If the pdf:stop-on-error flag is set to true, then Ghost-
script is requested to stop processing the PDF when the
first error is encountered. Otherwise it will attempt to
process all requested pages.
ps:imagemask
If the ps:imagemask flag is defined, the PS3 and EPS3
coders will create Postscript files that render bilevel
images with the Postscript imagemask operator instead of
the image operator.
ptif:minimum-geometry=<geometry>
If the ptif:minimum-geometry key is defined, GraphicsMag-
ick will use it to determine the minimum frame size to
output when writing a pyramid TIFF file (a TIFF file con-
taining a succession of reduced versions of the first
frame). The default minimum geometry is 32x32.
tiff:alpha={unspecified|associated|unassociated}
Specify the TIFF alpha channel type when reading or writ-
ing TIFF files, overriding the normal value. The default
alpha channel type for new files is unspecified alpha. Ex-
isting alpha settings are preserved when converting from
one TIFF file to another. When a TIFF file uses associated
alpha, the image pixels are pre-multiplied (i.e. altered)
with the alpha channel. Files with "associated" alpha ap-
pear as if they were alpha composited on a black back-
ground when the matte channel is disabled. If the unasso-
ciated alpha type is selected, then the alpha channel is
saved without altering the pixels. Photoshop recognizes
associated alpha as transparency information, if the file
is saved with unassociated alpha, the alpha information is
loaded as an independent channel. Note that for many
years, ImageMagick and GraphicsMagick marked TIFF files as
using associated alpha, without properly pre-multiplying
the pixels.
tiff:fill-order={msb2lsb|lsb2msb}
If the tiff:fill-order key is defined, GraphicsMagick will
use it to determine the bit fill order used while writing
TIFF files. The normal default is "msb2lsb", which matches
the native bit order of all modern CPUs. The only excep-
tion to this is when Group3 or Group4 FAX compression is
requested since FAX machines send data in bit-reversed or-
der and therefore RFC 2301 recommends using reverse order.
tiff:group-three-options=<value>
If the tiff:group-three-options key is defined, Graphics-
Magick will use it to set the group3 options tag when
writing group3-compressed TIFF. Please see the TIFF spec-
ification for the usage of this tag. The default value is
4.
tiff:ignore-tags=<tags>
If the tiff:ignore-tags key is defined, then it is used as
a list of comma-delimited integer TIFF tag values to ig-
nore while reading the TIFF file. This is useful in order
to be able to read files which which otherwise fail to
read due to problems with TIFF tags. Note that some TIFF
tags are required in order to be able to read the image
data at all.
tiff:report-warnings={false|true}
If the tiff:report-warnings key is defined and set to
true, then TIFF warnings are reported as a warning excep-
tion rather than as a coder log message. Such warnings
are reported after the image has been read or written.
Most TIFF warnings are benign but sometimes they may help
deduce problems with the TIFF file, or help detect that
the TIFF file requires a special application to read suc-
cessfully due to the use of proprietary or specialized ex-
tensions.
tiff:sample-format={unsigned|ieeefp}
If the tiff:sample-format key is defined, GraphicsMagick
will use it to determine the sample format used while
writing TIFF files. The default is "unsigned". Specify
"ieeefp" in order to write floating-point TIFF files with
float (32-bit) or double (64-bit) values. Use the
tiff:bits-per-sample define to determine the type of
floating-point value to use.
tiff:max-sample-value=<value>
If the tiff:max-sample-value key is defined, GraphicsMag-
ick will use the assigned value as the maximum floating
point value while reading or writing IEEE floating point
TIFFs. Otherwise the maximum value is 1.0 or the value ob-
tained from the file's SMaxSampleValue tag (if present).
The floating point data is currently not scanned in ad-
vance to determine a best maximum sample value so if the
range is not 1.0, or the SMaxSampleValue tag is not
present, it may be necessary to (intelligently) use this
parameter to properly read a file.
tiff:min-sample-value=<value>
If the tiff:min-sample-value key is defined, GraphicsMag-
ick will use the assigned value as the minimum floating
point value while reading or writing IEEE floating point
TIFFs. Otherwise the minimum value is 0.0 or the value ob-
tained from the file's SMinSampleValue tag (if present).
tiff:bits-per-sample=<value>
If the tiff:bits-per-sample key is defined, GraphicsMagick
will write images with the specified bits per sample,
overriding any existing depth value. Value may be any in
the range of 1 to 32, or 64 when the default ´unsigned'
format is written, or 16/32/24/64 if IEEEFP format is
written. Please note that the baseline TIFF 6.0 specifi-
cation only requires readers to handle certain powers of
two, and the values to be handled depend on the nature of
the image (e.g. colormapped, grayscale, RGB, CMYK).
tiff:samples-per-pixel=<value>
If the tiff:samples-per-pixel key is defined to a value,
the TIFF coder will write TIFF images with the defined
samples per pixel, overriding any value stored in the im-
age. This option should not normally be used.
tiff:rows-per-strip=<value>
Allows the user to specify the number of rows per TIFF
strip. Rounded up to a multiple of 16 when using JPEG
compression. Ignored when using tiles.
tiff:strip-per-page=true
Requests that the image is written in a single TIFF strip.
This is normally the default when group3 or group4 com-
pression is requested within reasonable limits. Requesting
a single strip for large images may result in failure due
to resource consumption in the writer or reader.
tiff:tile
Enable writing tiled TIFF (rather than stripped) using the
default tile size. Tiled TIFF organizes the image as an
array of smaller images (tiles) in order to enable random
access.
tiff:tile-geometry=<width>x<height>
Specify the tile size to use while writing tiled TIFF.
Width and height should be a multiple of 16. If the value
is not a multiple of 16, then it will be rounded down. En-
ables tiled TIFF if it has not already been enabled.
GraphicsMagick does not use tiled storage internally so
tiles need to be converted back and forth from the inter-
nal scanline-oriented storage to tile-oriented storage.
Testing with typical RGB images shows that useful square
tile size values range from 128x128 to 1024x1024. Large
images which require using a disk-based pixel cache bene-
fit from large tile sizes while images which fit in memory
work well with smaller tile sizes.
tiff:tile-width=<width>
Specify the tile width to use while writing tiled TIFF.
The tile height is then defaulted to an appropriate size.
Width should be a multiple of 16. If the value is not a
multiple of 16, then it will be rounded down. Enables
tiled TIFF if it has not already been enabled.
tiff:tile-height=<height>
Specify the tile height to use while writing tiled TIFF.
The tile width is then defaulted to an appropriate size.
Height should be a multiple of 16. If the value is not a
multiple of 16, then it will be rounded down. Enables
tiled TIFF if it has not already been enabled.
tiff:webp-lossless={TRUE|FALSE}
Specify a value of TRUE to enable lossless mode while
writing WebP-compressed TIFF files. The WebP webp:lossless
option may also be used. The quality factor set by the
-quality option may be used to influence the level of ef-
fort expended while compressing.
tiff:zstd-compress-level=<value>
Specify the compression level to use while writing Zstd-
compressed TIFF files. The valid range is 1 to 22. If this
define is not specified, then the 'quality' value is used
such that the default quality setting of 75 is translated
to a compress level of 9 such that ´quality' has a useful
range of 10-184 if used for this purpose.
webp:lossless={true|false}
Enable lossless encoding.
webp:method={0-6}
Quality/speed trade-off.
webp:image-hint={default,graph,photo,picture}
Hint for image type.
webp:target-size=<integer>
Target size in bytes.
webp:target-psnr=<float>
Minimal distortion to try to achieve.
webp:segments={1-4}
Maximum number of segments to use.
webp:sns-strength={0-100}
Spatial Noise Shaping.
webp:filter-strength={0-100}
Filter strength.
webp:filter-sharpness={0-7}
Filter sharpness.
webp:filter-type={0,1}
Filtering type. 0 = simple, 1 = strong (only used if fil-
ter-strength > 0 or autofilter is enabled).
webp:auto-filter={true|false}
Auto adjust filter's strength.
webp:alpha-compression=<integer>
Algorithm for encoding the alpha plane (0 = none, 1 = com-
pressed with WebP lossless). Default is 1.
webp:alpha-filtering=<integer>
Predictive filtering method for alpha plane. 0: none, 1:
fast, 2: best. Default is 1.
webp:alpha-quality={0-100}
Between 0 (smallest size) and 100 (lossless). Default is
100.
webp:pass=[1..10]
Number of entropy-analysis passes.
webp:show-compressed={true|false}
Export the compressed picture back. In-loop filtering is
not applied.
webp:preprocessing=[0,1,2]
0=none, 1=segment-smooth, 2=pseudo-random dithering
webp:partitions=[0-3]
log2(number of token partitions) in [0..3]. Default is 0
for easier progressive decoding.
webp:partition-limit={0-100}
Quality degradation allowed to fit the 512k limit on pre-
diction modes coding (0: no degradation, 100: maximum pos-
sible degradation).
webp:emulate-jpeg-size={true|false}
If true, compression parameters will be remapped to better
match the expected output size from JPEG compression. Gen-
erally, the output size will be similar but the degrada-
tion will be lower.
webp:thread-level=<integer>
If non-zero, try and use multi-threaded encoding.
webp:low-memory={true|false}
If set, reduce memory usage (but increase CPU use)
webp:use-sharp-yuv={true|false}
If set, if needed, use sharp (and slow) RGB->YUV conver-
sion
For example, to create a postscript file that will render only
the black pixels of a bilevel image, use:
gm convert bilevel.tif -define ps:imagemask eps3:stencil.ps
-delay <1/100ths of a second>
display the next image after pausing
This option is useful for regulating the animation of image se-
quences Delay/100 seconds must expire before the display of the
next image. The default is no delay between each showing of the
image sequence. The maximum delay is 65535.
You can specify a delay range (e.g. -delay 10-500) which sets
the minimum and maximum delay.
-density <width>x<height>
horizontal and vertical resolution in pixels of the image This
option specifies the image resolution to store while encoding a
raster image or the canvas resolution while rendering (reading)
vector formats such as Postscript, PDF, WMF, and SVG into a
raster image. Image resolution provides the unit of measure to
apply when rendering to an output device or raster image. The
default unit of measure is in dots per inch (DPI). The -units
option may be used to select dots per centimeter instead.
The default resolution is 72 dots per inch, which is equivalent
to one point per pixel (Macintosh and Postscript standard). Com-
puter screens are normally 72 or 96 dots per inch while printers
typically support 150, 300, 600, or 1200 dots per inch. To de-
termine the resolution of your display, use a ruler to measure
the width of your screen in inches, and divide by the number of
horizontal pixels (1024 on a 1024x768 display). If the file
format supports it, this option may be used to update the stored
image resolution. Note that Photoshop stores and obtains image
resolution from a proprietary embedded profile. If this profile
is not stripped from the image, then Photoshop will continue to
treat the image using its former resolution, ignoring the image
resolution specified in the standard file header. The density
option is an attribute and does not alter the underlying raster
image. It may be used to adjust the rendered size for desktop
publishing purposes by adjusting the scale applied to the pix-
els. To resize the image so that it is the same size at a dif-
ferent resolution, use the -resample option.
-depth <value>
depth of the image
This is the number of bits of color to preserve in the image.
Any value between 1 and QuantumDepth (build option) may be spec-
ified, although 8 or 16 are the most common values. Use this op-
tion to specify the depth of raw images whose depth is unknown
such as GRAY, RGB, or CMYK, or to change the depth of any image
after it has been read. The depth option is applied to the pix-
els immediately so it may be used as a form of simple compres-
sion by discarding the least significant bits. Reducing the
depth in advance may speed up color quantization, and help cre-
ate smaller file sizes when using a compression algorithm like
LZW or ZIP.
-descend
obtain image by descending window hierarchy
-despeckle
reduce the speckles within an image
-displace <horizontal scale>x<vertical scale>
shift image pixels as defined by a displacement map
With this option, composite image is used as a displacement map.
Black, within the displacement map, is a maximum positive dis-
placement. White is a maximum negative displacement and middle
gray is neutral. The displacement is scaled to determine the
pixel shift. By default, the displacement applies in both the
horizontal and vertical directions. However, if you specify
mask, composite image is the horizontal X displacement and mask
the vertical Y displacement.
-display <host:display[.screen]>
specifies the X server to contact
This option is used with convert for obtaining image or font
from this X server. See X(1).
-dispose <method>
GIF disposal method
The Disposal Method indicates the way in which the graphic is to
be treated after being displayed.
Here are the valid methods:
Undefined No disposal specified.
None Do not dispose between frames.
Background Overwrite the image area with
the background color.
Previous Overwrite the image area with
what was there prior to rendering
the image.
-dissolve <percent>
dissolve an image into another by the given percent
The opacity of the composite image is multiplied by the given
percent, then it is composited over the main image.
-dither
apply Floyd/Steinberg error diffusion to the image
The basic strategy of dithering is to trade intensity resolution
for spatial resolution by averaging the intensities of several
neighboring pixels. Images which suffer from severe contouring
when reducing colors can be improved with this option.
The -colors or -monochrome option is required for this option to
take effect.
Use +dither to turn off dithering and to render PostScript with-
out text or graphic aliasing. Disabling dithering often (but not
always) leads to decreased processing time.
-draw <string>
annotate an image with one or more graphic primitives
Use this option to annotate an image with one or more graphic
primitives. The primitives include shapes, text, transforma-
tions, and pixel operations. The shape primitives are
point x,y
line x0,y0 x1,y1
rectangle x0,y0 x1,y1
roundRectangle x0,y0 x1,y1 wc,hc
arc x0,y0 x1,y1 a0,a1
ellipse x0,y0 rx,ry a0,a1
circle x0,y0 x1,y1
polyline x0,y0 ... xn,yn
polygon x0,y0 ... xn,yn
Bezier x0,y0 ... xn,yn
path path specification
image operator x0,y0 w,h filename
The text primitive is
text x0,y0 string
The text gravity primitive is
gravity NorthWest, North, NorthEast, West, Center,
East, SouthWest, South, or SouthEast
The text gravity primitive only affects the placement of text
and does not interact with the other primitives. It is equiva-
lent to using the -gravity commandline option, except that it is
limited in scope to the -draw option in which it appears.
The transformation primitives are
rotate degrees
translate dx,dy
scale sx,sy
skewX degrees
skewY degrees
The pixel operation primitives are
color x0,y0 method
matte x0,y0 method
The shape primitives are drawn in the color specified in the
preceding -stroke option. Except for the line and point primi-
tives, they are filled with the color specified in the preceding
-fill option. For unfilled shapes, use -fill none.
Point requires a single coordinate.
Line requires a start and end coordinate.
Rectangle expects an upper left and lower right coordinate.
RoundRectangle has the upper left and lower right coordinates
and the width and height of the corners.
Circle has a center coordinate and a coordinate for the outer
edge.
Use Arc to inscribe an elliptical arc within a rectangle. Arcs
require a start and end point as well as the degree of rotation
(e.g. 130,30 200,100 45,90).
Use Ellipse to draw a partial ellipse centered at the given
point with the x-axis and y-axis radius and start and end of arc
in degrees (e.g. 100,100 100,150 0,360).
Finally, polyline and polygon require three or more coordinates
to define its boundaries. Coordinates are integers separated by
an optional comma. For example, to define a circle centered at
100,100 that extends to 150,150 use:
-draw 'circle 100,100 150,150'
Paths (See Paths) represent an outline of an object which is de-
fined in terms of moveto (set a new current point), lineto (draw
a straight line), curveto (draw a curve using a cubic Bezier),
arc (elliptical or circular arc) and closepath (close the cur-
rent shape by drawing a line to the last moveto) elements. Com-
pound paths (i.e., a path with subpaths, each consisting of a
single moveto followed by one or more line or curve operations)
are possible to allow effects such as "donut holes" in objects.
Use image to composite an image with another image. Follow the
image keyword with the composite operator, image location, image
size, and filename:
-draw 'image Over 100,100 225,225 image.jpg'
You can use 0,0 for the image size, which means to use the ac-
tual dimensions found in the image header. Otherwise, it will be
scaled to the given dimensions. See -compose for a description
of the composite operators.
Use text to annotate an image with text. Follow the text coordi-
nates with a string. If the string has embedded spaces, enclose
it in single or double quotes. Optionally you can include the
image filename, type, width, height, or other image attribute by
embedding special format character. See -comment for details.
For example,
-draw 'text 100,100 "%m:%f %wx%h"'
annotates the image with MIFF:bird.miff 512x480 for an image titled
bird.miff
and whose width is 512 and height is 480.
If the first character of string is @, the text is read
from a file titled by the remaining characters in the string. Please
note that if the string comes from an untrusted source that it should
be sanitized before use (a security risk).
Rotate rotates subsequent shape primitives and text primitives about
the origin of the main image. If the -region option precedes the
-draw option, the origin for transformations is the upper left
corner of the region.
Translate translates them.
Scale scales them.
SkewX and SkewY skew them with respect to the origin of
the main image or the region.
The transformations modify the current affine matrix, which is initialized
from the initial affine matrix defined by the -affine option.
Transformations are cumulative within the -draw option.
The initial affine matrix is not affected; that matrix is only changed by the
appearance of another -affine option. If another -draw
option appears, the current affine matrix is reinitialized from
the initial affine matrix.
Use color to change the color of a pixel to the fill color (see
-fill). Follow the pixel coordinate
with a method:
point
replace
floodfill
filltoborder
reset
Consider the target pixel as that specified by your coordinate.
The point method recolors the target pixel. The replace method
recolors any pixel that matches the color of the target pixel.
Floodfill recolors any pixel that matches the color of the tar-
get pixel and is a neighbor, whereas filltoborder recolors any
neighbor pixel that is not the border color. Finally, reset re-
colors all pixels.
Use matte to the change the pixel matte value to transparent.
Follow the pixel coordinate with a method (see the color primi-
tive for a description of methods). The point method changes the
matte value of the target pixel. The replace method changes the
matte value of any pixel that matches the color of the target
pixel. Floodfill changes the matte value of any pixel that
matches the color of the target pixel and is a neighbor, whereas
filltoborder changes the matte value of any neighbor pixel that
is not the border color (-bordercolor). Finally reset changes
the matte value of all pixels.
You can set the primitive color, font, and font bounding box
color with -fill, -font, and -box respectively. Options are pro-
cessed in command line order so be sure to use these options be-
fore the -draw option.
-edge <radius>
detect edges within an image
-emboss <radius>
emboss an image
-encoding <type>
specify the text encoding
Choose from AdobeCustom, AdobeExpert, AdobeStandard, AppleRoman,
BIG5, GB2312, Latin 2, None, SJIScode, Symbol, Unicode, Wansung.
-endian <type>
specify endianness (MSB, LSB, or Native) of image
MSB indicates big-endian (e.g. SPARC, Motorola 68K) while LSB
indicates little-endian (e.g. Intel 'x86, VAX) byte ordering.
Native indicates to use the normal ordering for the current CPU.
This option currently only influences the CMYK, DPX, GRAY, RGB,
and TIFF, formats.
Use +endian to revert to unspecified endianness.
-enhance
apply a digital filter to enhance a noisy image
-equalize
perform histogram equalization to the image
-extent <width>x<height>{+-}<x>{+-}<y>
composite image on background color canvas image
This option composites the image on a new background color
(-background) canvas image of size <width>x<height>. The exist-
ing image content is composited at the position specified by ge-
ometry x and y offset and/or desired gravity (-gravity) using
the current image compose (-compose) method. Image content
which falls outside the bounds of the new image dimensions is
discarded.
For example, this command creates a thumbnail of an image, and
centers it on a red color backdrop image, offsetting the canvas
ten pixels to the left and five pixels up, with respect to the
thumbnail:
gm convert infile.jpg -thumbnail 120x80 -background red -gravity center \
-extent 140x100-10-5 outfile.jpg
This command reduces or expands a JPEG image to fit on an
800x600 display:
gm convert -size 800x600 input.jpg \
-resize 800x600 -background black \
-compose Copy -gravity center \
-extent 800x600 \
-quality 92 output.jpg
If the aspect ratio of the input image isn't exactly 4:3, then
the image is centered on an 800x600 black canvas.
-file <filename>
write annotated difference image to file
If -file is specified, then an annotated difference image is
generated and written to the specified file. Pixels which differ
between the reference and compare images are modified from those
in the compare image so that the changed pixels become more ob-
vious. Some images may require use of an alternative highlight
style (see -highlight-style) or highlight color (see -highlight-
color) before the changes are obvious.
-fill <color>
color to use when filling a graphic primitive
Colors are represented in GraphicsMagick in the same form used
by SVG. Use "gm convert -list color" to list named colors:
name (named color)
#RGB (hex numbers, 4 bits each)
#RRGGBB (8 bits each)
#RRRGGGBBB (12 bits each)
#RRRRGGGGBBBB (16 bits each)
#RGBA (4 bits each)
#RRGGBBAA (8 bits each)
#RRRGGGBBBAAA (12 bits each)
#RRRRGGGGBBBBAAAA (16 bits each)
rgb(r,g,b) (r,g,b are decimal numbers)
rgba(r,g,b,a) (r,g,b,a are decimal numbers)
Enclose the color specification in quotation marks to prevent
the "#" or the parentheses from being interpreted by your shell.
For example,
gm convert -fill blue ...
gm convert -fill "#ddddff" ...
gm convert -fill "rgb(65000,65000,65535)" ...
The shorter forms are scaled up, if necessary by replication.
For example, #3af, #33aaff, and #3333aaaaffff are all equiva-
lent.
See -draw for further details.
-filter <type>
use this type of filter when resizing an image
Use this option to affect the resizing operation of an image
(see -geometry). Choose from these filters (ordered by approxi-
mate increasing CPU time):
Point
Box
Triangle
Hermite
Hanning
Hamming
Blackman
Gaussian
Quadratic
Cubic
Catrom
Mitchell
Lanczos
Bessel
Sinc
The default filter is automatically selected to provide the best
quality while consuming a reasonable amount of time. The
Mitchell filter is used if the image supports a palette, sup-
ports a matte channel, or is being enlarged, otherwise the Lanc-
zos filter is used.
-flatten
flatten a sequence of images
In some file formats (e.g. Photoshop's PSD) complex images may
be represented by "layers" (independent images) which must be
composited in order to obtain the final rendition. The -flatten
option accomplishes this composition. The sequence of images is
replaced by a single image created by compositing each image in
turn, while respecting composition operators and page offsets.
While -flatten is immediately useful for eliminating layers, it
is also useful as a general-purpose composition tool.
The sequence of images is terminated by the appearance of any
option. If the -flatten option appears after all of the input
images, all images are flattened. Also see -mosaic which is
similar to -flatten except that it adds a suitably-sized canvas
base image.
For example, this composites an image on top of a 640x400 trans-
parent black canvas image:
gm convert -size 640x300 xc:transparent \
-compose over -page +0-100 \
frame.png -flatten output.png
and this flattens a Photoshop PSD file:
gm convert input.psd -flatten output.png
-flip create a "mirror image"
reflect the scanlines in the vertical direction.
-flop create a "mirror image"
reflect the scanlines in the horizontal direction.
-font <name>
use this font when annotating the image with text
You can tag a font to specify whether it is a PostScript, True-
Type, or X11 font. For example, Arial.ttf is a TrueType font,
ps:helvetica is PostScript, and x:fixed is X11.
-foreground <color>
define the foreground color
The color is specified using the format described under the
-fill option.
-format <type>
the image format type
When used with the mogrify utility, this option will convert any
image to the image format you specify. See GraphicsMagick(1)
for a list of image format types supported by GraphicsMagick, or
see the output of 'gm -list format'.
By default the file is written to its original name. However,
if the filename extension matches a supported format, the exten-
sion is replaced with the image format type specified with -for-
mat. For example, if you specify tiff as the format type and
the input image filename is image.gif, the output image filename
becomes image.tiff.
-format <string>
output formatted image characteristics
When used with the identify utility, or the convert utility with
output written to the 'info:-' file specification, use this op-
tion to print information about the image in a format of your
choosing. You can include the image filename, type, width,
height, Exif data, or other image attributes by embedding spe-
cial format characters:
%b file size
%c comment
%d directory
%e filename extension
%f filename
%g page dimensions and offsets
%h height
%i input filename
%k number of unique colors
%l label
%m magick
%n number of scenes
%o output filename
%p page number
%q image bit depth
%r image type description
%s scene number
%t top of filename
%u unique temporary filename
%w width
%x horizontal resolution
%y vertical resolution
%A transparency supported
%C compression type
%D GIF disposal method
%G Original width and height
%H page height
%M original filename specification
%O page offset (x,y)
%P page dimensions (width,height)
%Q compression quality
%T time delay (in centi-seconds)
%U resolution units
%W page width
%X page horizontal offset (x)
%Y page vertical offset (y)
%@ trim bounding box
%# signature
\n newline
\r carriage return
%% %
For example,
-format "%m:%f %wx%h"
displays MIFF:bird.miff 512x480 for an image titled bird.miff
and whose width is 512 and height is 480.
If the first character of string is @, the format is read from a
file titled by the remaining characters in the string. Please
note that if the string comes from an untrusted source that it
should be sanitized before use since this may be used to incor-
porate any readable file on the system (a security risk).
The values of image type (%r) which may be returned include:
Bilevel
Grayscale
GrayscaleMatte
Palette
PaletteMatte
TrueColor
TrueColorMatte
ColorSeparation
ColorSeparationMatte
Optimize
You can also use the following special formatting syntax to
print Exif information contained in the file:
%[EXIF:<tag>]
Where "<tag>" may be one of the following:
* (print all Exif tags, in keyword=data format)
! (print all Exif tags, in tag_number format)
#hhhh (print data for Exif tag #hhhh)
ImageWidth
ImageLength
BitsPerSample
Compression
PhotometricInterpretation
FillOrder
DocumentName
ImageDescription
Make
Model
StripOffsets
Orientation
SamplesPerPixel
RowsPerStrip
StripByteCounts
XResolution
YResolution
PlanarConfiguration
ResolutionUnit
TransferFunction
Software
DateTime
Artist
WhitePoint
PrimaryChromaticities
TransferRange
JPEGProc
JPEGInterchangeFormat
JPEGInterchangeFormatLength
YCbCrCoefficients
YCbCrSubSampling
YCbCrPositioning
ReferenceBlackWhite
CFARepeatPatternDim
CFAPattern
BatteryLevel
Copyright
ExposureTime
FNumber
IPTC/NAA
ExifOffset
InterColorProfile
ExposureProgram
SpectralSensitivity
GPSInfo
ISOSpeedRatings
OECF
ExifVersion
DateTimeOriginal
DateTimeDigitized
ComponentsConfiguration
CompressedBitsPerPixel
ShutterSpeedValue
ApertureValue
BrightnessValue
ExposureBiasValue
MaxApertureValue
SubjectDistance
MeteringMode
LightSource
Flash
FocalLength
MakerNote
UserComment
SubSecTime
SubSecTimeOriginal
SubSecTimeDigitized
FlashPixVersion
ColorSpace
ExifImageWidth
ExifImageLength
InteroperabilityOffset
FlashEnergy
SpatialFrequencyResponse
FocalPlaneXResolution
FocalPlaneYResolution
FocalPlaneResolutionUnit
SubjectLocation
ExposureIndex
SensingMethod
FileSource
SceneType
JPEG specific information (from reading a JPEG file) may be ob-
tained like this:
%[JPEG-<tag>]
Where "<tag>" may be one of the following:
* (all JPEG-related tags, in
keyword=data format)
Quality IJG JPEG "quality" estimate
Colorspace JPEG colorspace numeric ID
Colorspace-Name JPEG colorspace name
Sampling-factors JPEG sampling factors
Please note that JPEG has no notion of "quality" and that the
quality metric used by, and estimated by the software is based
on the quality metric established by IJG JPEG 6b. Other en-
coders (e.g. that used by Adobe Photoshop) use different encod-
ing metrics.
Surround the format specification with quotation marks to pre-
vent your shell from misinterpreting any spaces and square
brackets.
-frame <width>x<height>+<outer bevel width>+<inner bevel width>
surround the image with an ornamental border
See -geometry for details about the geometry specification. The
-frame option is not affected by the -gravity option.
The color of the border is specified with the -mattecolor com-
mand line option.
-frame include the X window frame in the imported image
-fuzz <distance>{%}
colors within this Euclidean distance are considered equal
A number of algorithms search for a target color. By default the
color must be exact. Use this option to match colors that are
close (in Euclidean distance) to the target color in RGB 3D
space. For example, if you want to automatically trim the edges
of an image with -trim but the image was scanned and the target
background color may differ by a small amount. This option can
account for these differences.
The distance can be in absolute intensity units or, by appending
"%", as a percentage of the maximum possible intensity (255,
65535, or 4294967295).
-gamma <value>
level of gamma correction
The same color image displayed on two different workstations may
look different due to differences in the display monitor. Use
gamma correction to adjust for this color difference. Reasonable
values extend from 0.8 to 2.3. Gamma less than 1.0 darkens the
image and gamma greater than 1.0 lightens it. Large adjustments
to image gamma may result in the loss of some image information
if the pixel quantum size is only eight bits (quantum range 0 to
255).
You can apply separate gamma values to the red, green, and blue
channels of the image with a gamma value list delimited with
slashes (e.g., 1.7/2.3/1.2).
Use +gamma value to set the image gamma level without actually
adjusting the image pixels. This option is useful if the image
is of a known gamma but not set as an image attribute (e.g. PNG
images).
-gaussian <radius>{x<sigma>}
blur the image with a Gaussian operator
Use the given radius and standard deviation (sigma).
-geometry <width>x<height>{+-}<x>{+-}<y>{%}{@}{!}{^}{<}{>}
Specify dimension, offset, and resize options.
The -geometry option is used for a number of different purposes,
depending on the utility it is used with.
For the X11 commands ('animate', 'display', and 'import'), it
specifies the preferred size and location of the Image window.
By default, the window size is the image size and the location
is chosen by you (or your window manager) when it is mapped.
For the 'import', 'convert', 'mogrify' utility commands it may
be used to specify the desired size when resizing an image. In
this case, symbols representing resize options may be appended
to the geometry string to influence how the resize request is
treated.
See later notes corresponding to usage by particular commands.
The following notes apply to when -geometry is used to express a
resize request, taking into account the current properties of
the image.
By default, the width and height are maximum values. That is,
the image is expanded or contracted to fit the width and height
value while maintaining the aspect ratio of the image.
Append a ^ to the geometry so that the image aspect ratio is
maintained when the image is resized, but the resulting width or
height are treated as minimum values rather than maximum values.
Append a ! (exclamation point) to the geometry to force the im-
age size to exactly the size you specify. For example, if you
specify 640x480! the image width is set to 640 pixels and height
to 480.
If only the width is specified, without the trailing 'x', then
height is set to width (e.g., -geometry 100 is the same as -ge-
ometry 100x100). If only the width is specified but with the
trailing 'x', then width assumes the value and the height is
chosen to maintain the aspect ratio of the image. Similarly, if
only the height is specified prefixed by 'x' (e.g., -geometry
x256), the width is chosen to maintain the aspect ratio.
To specify a percentage width or height instead, append %. The
image size is multiplied by the width and height percentages to
obtain the final image dimensions. To increase the size of an
image, use a value greater than 100 (e.g. 125%). To decrease an
image's size, use a percentage less than 100.
Use @ to specify the maximum area in pixels of an image.
Use > to change the dimensions of the image only if its width or
height exceeds the geometry specification. < resizes the image
only if both of its dimensions are less than the geometry speci-
fication. For example, if you specify '640x480>' and the image
size is 256x256, the image size does not change. However, if the
image is 512x512 or 1024x1024, it is resized to 480x480. En-
close the geometry specification in quotation marks to prevent
the < or > from being interpreted by your shell as a file redi-
rection.
When used with animate and display, offsets are handled in the
same manner as in X(1) and the -gravity option is not used. If
the x is negative, the offset is measured leftward from the
right edge of the screen to the right edge of the image being
displayed. Similarly, negative y is measured between the bottom
edges. The offsets are not affected by "%"; they are always
measured in pixels.
When used as a composite option, -geometry gives the dimensions
of the image and its location with respect to the composite im-
age. If the -gravity option is present with NorthEast, East, or
SouthEast gravity, the x represents the distance from the right
edge of the image to the right edge of the composite image.
Similarly, if the -gravity option is present with SouthWest,
South, or SouthEast gravity, y is measured between the bottom
edges. Accordingly, a positive offset will never point in the
direction outside of the image. The offsets are not affected by
"%"; they are always measured in pixels. To specify the dimen-
sions of the composite image, use the -resize option.
When used as a convert, import or mogrify option, -geometry is
synonymous with -resize and specifies the size of the output im-
age. The offsets, if present, are ignored.
When used as a montage option, -geometry specifies the image
size and border size for each tile; default is 256x256+0+0.
Negative offsets (border dimensions) are meaningless. The
-gravity option affects the placement of the image within the
tile; the default gravity for this purpose is Center. If the
"%" sign appears in the geometry specification, the tile size is
the specified percentage of the original dimensions of the first
tile. To specify the dimensions of the montage, use the -resize
option.
-gravity <type>
direction primitive gravitates to when annotating the image.
Choices are: NorthWest, North, NorthEast, West, Center, East,
SouthWest, South, SouthEast.
The direction you choose specifies where to position the text
when annotating the image. For example Center gravity forces the
text to be centered within the image. By default, the image
gravity is NorthWest. See -draw for more details about graphic
primitives. Only the text primitive is affected by the -gravity
option.
The -gravity option is also used in concert with the -geometry
option and other options that take <geometry> as a parameter,
such as the -crop option. See -geometry for details of how the
-gravity option interacts with the <x> and <y> parameters of a
geometry specification.
When used as an option to composite, -gravity gives the direc-
tion that the image gravitates within the composite.
When used as an option to montage, -gravity gives the direction
that an image gravitates within a tile. The default gravity is
Center for this purpose.
-green-primary <x>,<y>
green chromaticity primary point
-hald-clut <clut>
apply a Hald CLUT to the image
A Hald CLUT ("Color Look-Up Table") is a special square color
image which contains a look-up table for red, green, and blue.
The size of the Hald CLUT image is determined by its order. The
width (and height) of a Hald CLUT is the cube of the order. For
example, a Hald CLUT of order 8 is 512x512 pixels (262,144 col-
ors) and of order 16 is 4096x4096 (16,777,216 colors). A spe-
cial CLUT is the identity CLUT which which causes no change to
the input image. In order to use the Hald CLUT, one takes an
identity CLUT and adjusts its colors in some way. The modified
CLUT can then be used to transform any number of images in an
identical way.
GraphicsMagick contains a built-in identity CLUT generator via
the IDENTITY coder. For example reading from the file name
IDENTITY:8 returns an identity CLUT of order 8. Typical Hald
CLUT identity images have an order of between 8 and 16. The de-
fault order for the IDENTITY CLUT generator is 8. Interpolation
is used so it is not usually necessary for CLUT images to be
very large. The PNG file format is ideal for storing Hald CLUT
images because it compresses them very well.
-help print usage instructions
-highlight-color <color>
pixel annotation color
Specifies the color to use when annotating difference pixels.
-highlight-style <style>
pixel annotation style
Specifies the pixel difference annotation style used to draw at-
tention to changed pixels. May be one of Assign, Threshold,
Tint, or XOR; where Assign replaces the pixel with the highlight
color (see -highlight-color), Threshold replaces the pixel with
black or white based on the difference in intensity, Tint alpha
tints the pixel with the highlight color, and XOR does an XOR
between the pixel and the highlight color.
-iconGeometry <geometry>
specify the icon geometry
Offsets, if present in the geometry specification, are handled
in the same manner as the -geometry option, using X11 style to
handle negative offsets.
-iconic
iconic animation
-immutable
make image immutable
-implode <factor>
implode image pixels about the center
-intent <type>
use this type of rendering intent when managing the image color
Use this option to affect the the color management operation of
an image (see -profile). Choose from these intents: Absolute,
Perceptual, Relative, Saturation.
The default intent is undefined.
-interlace <type>
the type of interlacing scheme
Choices are: None, Line, Plane, or Partition. The default is
None.
This option is used to specify the type of interlacing scheme
for raw image formats such as RGB or YUV. None means do not in-
terlace (RGBRGBRGBRGBRGBRGB...),
Line uses scanline interlacing
(RRR...GGG...BBB...RRR...GGG...BBB...), and Plane uses plane in-
terlacing (RRRRRR...GGGGGG...BBBBBB...).
Partition is like plane except the different planes are saved to
individual files (e.g. image.R, image.G, and image.B).
Use Line to create an interlaced PNG or GIF or progressive JPEG
image.
-label <name>
assign a label to an image
Use this option to assign a specific label to the image, when
writing to an image format that supports labels, such as TIFF,
PNG, MIFF, or PostScript. You can include the the image file-
name, type, width, height, or other image attribute by embedding
special format character. A label is not drawn on the image,
but is embedded in the image datastream via a "Label" tag or
similar mechanism. If you want the label to be visible on the
image itself, use the -draw option. See -comment for details.
For example,
-label "%m:%f %wx%h"
produces an image label of MIFF:bird.miff 512x480 for an image
titled bird.miff and whose width is 512 and height is 480.
If the first character of string is @, the image label is read
from a file titled by the remaining characters in the string.
Please note that if the string comes from an untrusted source
that it should be sanitized before use since otherwise the con-
tent of an arbitrary readable file might be incorporated into
the image label (a security risk).
If the -label option appears multiple times, only the last label
is stored.
In PNG images, the label is stored in a tEXt or zTXt chunk with
the keyword "label".
When converting to PostScript, use this option to specify a
header string to print above the image. Specify the label font
with -font.
When creating a montage, by default the label associated with an
image is displayed with the corresponding tile in the montage.
Use the +label option to suppress this behavior.
-lat <width>x<height>{+-}<offset>{%}
perform local adaptive thresholding
Perform local adaptive thresholding using the specified width,
height, and offset. The offset is a distance in sample space
from the mean, as an absolute integer ranging from 0 to the max-
imum sample value or as a percentage. If the percent option is
supplied, then the offset is computed as a percentage of the
quantum range. It is strongly recommended to use the percent
option so that results are not sensitive to pixel quantum depth.
For example,
-colorspace gray -lat "10x10-5%"
will help clarify a scanned grayscale or color document, produc-
ing a bi-level equivalent.
-level <black_point>{,<gamma>}{,<white_point>}{%}
adjust the level of image contrast
Give one, two or three values delimited with commas: black-
point, gamma, white-point (e.g. 10,1.0,250 or 2%,0.5,98%). The
black and white points range from 0 to MaxRGB or from 0 to 100%;
if the white point is omitted it is set to MaxRGB-black_point.
If a "%" sign is present anywhere in the string, the black and
white points are percentages of MaxRGB. Gamma is an exponent
that ranges from 0.1 to 10.; if it is omitted, the default of
1.0 (no gamma correction) is assumed. This interface works simi-
lar to Photoshop's "Image->Adjustments->Levels..." "Input Lev-
els" interface.
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
By default, resource limits are estimated based on the available
resources and capabilities of the system. The resource limits
are Disk, maximum total disk space consumed; File, maximum num-
ber of file descriptors allowed to be open at once; Map, maximum
total number of file bytes which may be memory mapped; Memory,
maximum total number of bytes of heap memory used for image
storage; Pixels, maximum absolute image size (per image); Width,
maximum image pixels width; Height, maximum image pixels height;
Read, maximum number of uncompressed bytes to read; and Threads,
the maximum number of worker threads to use per OpenMP thread
team.
The Disk and Map resource limits are used to decide if (for a
given image) the decoded image ("pixel cache") should be stored
in heap memory (RAM), in a memory-mapped disk file, or in a disk
file accessed via read/write I/O.
The number of total pixels in one image (Pixels), and/or the
width/height (Width/Height), may be limited in order to force
the reading, or creation of images larger than the limit (in
pixels) to intentionally fail. The disk limit (Disk) establishes
an overall limit since using the disk is the means of last re-
sort. When the disk limit has been reached, no more images may
be read.
The amount of uncompressed data read when reading one image may
be limited by the Read limit. Reading the image fails when the
limit is hit. This option is useful if the data is read from a
stream (pipe) or from a compressed file such as a gzipped file.
Some files are very compressable and so a small compressed file
can decompress to a huge amount of data. This option also de-
fends against files which produce seemingly endless loops while
decoding by seeking backwards in the file.
The value argument is an absolute value, but may have standard
binary suffix characters applied ('K', 'M', 'G', 'T', 'P', 'E')
to apply a scaling to the value (based on a multiplier of 1024).
Any additional characters are ignored. For example, '-limit Pix-
els 10MP' limits the maximum image size to 10 megapixels and
'-limit memory 32MB -limit map 64MB' limits memory and memory
mapped files to 32 megabytes and 64 megabytes respectively.
Resource limits may also be set using environment variables. The
environment variables MAGICK_LIMIT_DISK, MAGICK_LIMIT_FILES,
MAGICK_LIMIT_MAP, MAGICK_LIMIT_MEMORY, MAGICK_LIMIT_PIXELS, MAG-
ICK_LIMIT_WIDTH, MAGICK_LIMIT_HEIGHT. MAGICK_LIMIT_READ, and
OMP_NUM_THREADS may be used to set the limits for disk space,
open files, memory mapped size, heap memory, per-image pixels,
image width, image height, and threads respectively.
Use the option -list resource list the current limits.
-linewidth
the line width for subsequent draw operations
-list <type>
the type of list
Choices are: Color, Delegate, Format, Magic, Module, Resource,
or Type. The Module option is only available if GraphicsMagick
was built to support loadable modules.
This option lists information about the GraphicsMagick configu-
ration.
-log <string>
Specify format for debug log
This option specifies the format for the log printed when the
-debug option is active.
You can display the following components by embedding special
format characters:
%d domain
%e event
%f function
%l line
%m module
%p process ID
%r real CPU time
%t wall clock time
%u user CPU time
%% percent sign
\n newline
\r carriage return
For example:
gm convert -debug coders -log "%u %m:%l %e" in.gif out.png
The default behavior is to print all of the components.
-loop <iterations>
add Netscape loop extension to your GIF animation
A value other than zero forces the animation to repeat itself up
to iterations times.
-magnify
magnify the image
The image size is doubled using linear interpolation.
-magnify <factor>
magnify the image
The displayed image is magnified by factor.
-map <filename>
choose a particular set of colors from this image
[convert or mogrify]
By default, color reduction chooses an optimal set of colors
that best represent the original image. Alternatively, you can
choose a particular set of colors from an image file with this
option.
Use +map to reduce all images in the image sequence that follows
to a single optimal set of colors that best represent all the
images. The sequence of images is terminated by the appearance
of any option. If the +map option appears after all of the in-
put images, all images are mapped.
-map <type>
display image using this type.
[animate or display]
Choose from these Standard Colormap types:
best
default
gray
red
green
blue
The X server must support the Standard Colormap you choose, oth-
erwise an error occurs. Use list as the type and display
searches the list of colormap types in top-to-bottom order until
one is located. See xstdcmap(1) for one way of creating Standard
Colormaps.
-mask <filename>
Specify a clipping mask
The image read from the file is used as a clipping mask. It
must have the same dimensions as the image being masked.
If the mask image contains an opacity channel, the opacity of
each pixel is used to define the mask. Otherwise, the intensity
(gray level) of each pixel is used. Unmasked (black) pixels are
modified while masked pixels (not black) are protected from al-
teration.
Use +mask to remove the clipping mask.
It is not necessary to use -clip to activate the mask; -clip is
implied by -mask.
-matte store matte channel if the image has one
If the image does not have a matte channel, create an opaque
one.
Use +matte to ignore the matte channel (treats it as opaque) and
to avoid writing a matte channel in the output file.
For the compare command, -matte will add an opaque matte channel
to images if they do not already have a matte channel, and matte
will be enabled for both images. Likewise, if +matte is used,
the matte channel is disabled for both images. This makes it
easier to compare images regardless of if they already have a
matte channel.
-mattecolor <color>
specify the color to be used with the -frame option
The color is specified using the format described under the
-fill option.
-maximum-error <limit>
specifies the maximum amount of total image error
Specifies the maximum amount of total image error (based on com-
parison using a specified metric) before an error ("image dif-
ference exceeds limit") is reported. The error is reported via
a non-zero command execution return status.
-median <radius>
apply a median filter to the image
-metric <metric>
comparison metric (MAE, MSE, PAE, PSNR, RMSE)
-minify <factor>
minify the image
The image size is halved using linear interpolation.
-mode <value>
mode of operation
The available montage modes are frame to place the images in a
rectangular grid while adding a decorative frame with drop-
shadow, unframe to place undecorated images in a rectangular
grid, and concatenate to pack the images closely together with-
out any well-defined grid or decoration.
-modulate brightness[,saturation[,hue]]
vary the brightness, saturation, and hue of an image
Specify the percent change in brightness, color saturation, and
hue separated by commas. Default argument values are 100 per-
cent, resulting in no change. For example, to increase the color
brightness by 20% and decrease the color saturation by 10% and
leave the hue unchanged, use: -modulate 120,90.
Hue is the percentage of absolute rotation from the current po-
sition. For example 50 results in a counter-clockwise rotation
of 90 degrees, 150 results in a clockwise rotation of 90 de-
grees, with 0 and 200 both resulting in a rotation of 180 de-
grees.
-monitor
show progress indication
A simple command-line progress indication is shown while the
command is running. The process indication shows the operation
currently being performed and the percent completed. Commands
using X11 may replace the command line progress indication with
a graphical one once an image has been displayed.
-monochrome
transform the image to black and white
-morph <frames>
morphs an image sequence
Both the image pixels and size are linearly interpolated to give
the appearance of a meta-morphosis from one image to the next.
The sequence of images is terminated by the appearance of any
option. If the -morph option appears after all of the input im-
ages, all images are morphed.
-mosaic
create a mosaic from an image or an image sequence
The -mosaic option provides a flexible way to composite one or
more images onto a solid-color canvas image. It works similar to
-flatten except that a base canvas image is automatically cre-
ated with a suitable size given the image size, page dimensions,
and page offsets of images to be composited. The color of the
base canvas image may be set via the -background option. The
default canvas color is 'white', but 'black' or 'transparent'
may be more suitable depending on the composition algorithm re-
quested.
The -compose option may be used to specify the composition algo-
rithm to use when compositing the subsequent image on the base
canvas.
The -page option can be used to establish the dimensions of the
mosaic and to position the subsequent image within the mosaic.
If the -page argument does not specify width and height, then
the canvas dimensions are evaluated based on the image sizes and
offsets.
The sequence of images is terminated by the appearance of any
option. If the -mosaic option appears after all of the input
images, all images are included in the mosaic.
The following is an example of composing an image based on red,
green, and blue layers extracted from a sequence of images and
pasted on the canvas image at specified offsets:
gm convert -background black \
-compose CopyRed -page +0-100 red.png \
-compose CopyGreen -page +0+40 green.png \
-compose CopyBlue -page +0+180 blue.png \
-mosaic output.png
-motion-blur <radius>{x<sigma>}{+angle}
Simulate motion blur
Simulate motion blur by convolving the image with a Gaussian op-
erator of the given radius and standard deviation (sigma). For
reasonable results, radius should be larger than sigma. If ra-
dius is zero, then a suitable radius is automatically selected
based on sigma. The angle specifies the angle that the object is
coming from (side which is blurred).
-name name an image
-negate
replace every pixel with its complementary color
The red, green, and blue intensities of an image are negated.
White becomes black, yellow becomes blue, etc. Use +negate to
only negate the grayscale pixels of the image.
-noise <radius|type>
add or reduce noise in an image
The principal function of noise peak elimination filter is to
smooth the objects within an image without losing edge informa-
tion and without creating undesired structures. The central idea
of the algorithm is to replace a pixel with its next neighbor in
value within a pixel window, if this pixel has been found to be
noise. A pixel is defined as noise if and only if this pixel is
a maximum or minimum within the pixel window.
Use radius to specify the width of the neighborhood.
Use +noise followed by a noise type to add noise to an image.
The noise added modulates the existing image pixels. Choose from
these noise types:
Uniform
Gaussian
Multiplicative
Impulse
Laplacian
Poisson
Random (uniform distribution)
-noop NOOP (no option)
The -noop option can be used to terminate a group of images and
reset all options to their default values, when no other option
is desired.
-normalize
transform image to span the full range of color values
This is a contrast enhancement technique based on the image his-
togram.
When computing the contrast enhancement values, the histogram
edges are truncated so that the majority of the image pixels are
considered in the constrast enhancement, and outliers (e.g. ran-
dom noise or minute details) are ignored. The default is that
0.1 percent of the histogram entries are ignored. The percent-
age of the histogram to ignore may be specified by using the
-set option with the histogram-threshold parameter similar to
-set histogram-threshold 0.01 to specify 0.01 percent. Use 0
percent to use the entire histogram, with possibly diminished
contrast enhancement.
-opaque <color>
change this color to the pen color within the image
The color is specified using the format described under the
-fill option. The color is replaced if it is identical to the
target color, or close enough to the target color in a 3D space
as defined by the Euclidean distance specified by -fuzz.
See -fill and -fuzz for more details.
-operator channel operator rvalue[%]
apply a mathematical, bitwise, or value operator to an image
channel
Apply a low-level mathematical, bitwise, or value operator to a
selected image channel or all image channels. Operations which
result in negative results are reset to zero, and operations
which overflow the available range are reset to the maximum pos-
sible value.
Select a channel from: Red, Green, Blue, Opacity, Matte, Cyan,
Magenta, Yellow, Black, All, or Gray. All only modifies the
color channels and does not modify the Opacity channel. Except
for the threshold operators, All operates on each channel inde-
pendently so that operations are on a per-channel basis.
Gray treats the color channels as a grayscale intensity and per-
forms the requested operation on the equivalent pixel intensity
so the result is a gray image. Select an operator from Add,
And, Assign, Depth, Divide, Gamma, Negate, LShift, Log, Max,
Min, Multiply, Or, Pow, RShift, Subtract, Threshold, Threshold-
White, Threshold-White-Negate, Threshold-Black, Threshold-Black-
Negate, Xor, Noise-Gaussian, Noise-Impulse, Noise-Laplacian,
Noise-Multiplicative, Noise-Poisson, Noise-Random, and Noise-
Uniform.
Rvalue may be any floating point or integer value. Normally
rvalue will be in the range of 0 to MaxRGB, where MaxRGB is the
largest quantum value supported by the GraphicsMagick build
(255, 65535, or 4294967295) but values outside this range are
useful for some arithmetic operations. Arguments to logical or
bit-wise operations are rounded to a positive integral value
prior to use. If a percent (%) symbol is appended to the argu-
ment, then the argument has a range of 0 to 100 percent.
The following is a description of the operators:
Add
Result is rvalue added to channel value.
And
Result is the logical AND of rvalue with channel value.
Assign
Result is rvalue.
Depth
Result is channel value adjusted so that it may be (ap-
proximately) stored in the specified number of bits with-
out additional loss.
Divide
Result is channel value divided by rvalue.
Gamma
Result is channel value gamma adjusted by rvalue.
LShift
Result is channel value bitwise left shifted by rvalue
bits.
Log
Result is computed as log(value*rvalue+1)/log(rvalue+1).
Max
Result is assigned to rvalue if rvalue is greater than
value.
Min
Result is assigned to rvalue if rvalue is less than value.
Multiply
Result is channel value multiplied by rvalue.
Negate
Result is inverse of channel value (like a film negative).
An rvalue must be supplied but is currently not used. In-
verting the image twice results in the original image.
Or
Result is the logical OR of rvalue with channel value.
Pow
Result is computed as pow(value,rvalue). Similar to Gamma
except that rvalue is not inverted.
RShift
Result is channel value bitwise right shifted by rvalue
bits.
Subtract
Result is channel value minus rvalue.
Threshold
Result is maximum (white) if channel value is greater than
rvalue, or minimum (black) if it is less than or equal to
rvalue. If all channels are specified, then thresholding
is done based on computed pixel intensity.
Threshold-white
Result is maximum (white) if channel value is greater than
rvalue and is unchanged if it is less than or equal to
rvalue. This can be used to remove apparent noise from the
bright parts of an image. If all channels are specified,
then thresholding is done based on computed pixel inten-
sity.
Threshold-White-Negate
Result is set to black if channel value is greater than
rvalue and is unchanged if it is less than or equal to
rvalue. If all channels are specified, then thresholding
is done based on computed pixel intensity.
Threshold-black
Result is minimum (black) if channel value is less than
than rvalue and is unchanged if it is greater than or
equal to rvalue. This can be used to remove apparent noise
from the dark parts of an image. If all channels are spec-
ified, then thresholding is done based on computed pixel
intensity.
Threshold-Black-Negate
Result is set to white if channel value is less than than
rvalue and is unchanged if it is greater than or equal to
rvalue. If all channels are specified, then thresholding
is done based on computed pixel intensity.
Xor
Result is the logical XOR of rvalue with channel value. An
interesting property of XOR is that performing the same
operation twice results in the original value.
Noise-Gaussian
Result is the current channel value modulated with gauss-
ian noise according to the intensity specified by rvalue.
Noise-Impulse
Result is the current channel value modulated with impulse
noise according to the intensity specified by rvalue.
Noise-Laplacian
Result is the current channel value modulated with lapla-
cian noise according to the intensity specified by rvalue.
Noise-Multiplicative
Result is the current channel value modulated with multi-
plicative gaussian noise according to the intensity speci-
fied by rvalue.
Noise-Poisson
Result is the current channel value modulated with poisson
noise according to the intensity specified by rvalue.
Noise-Random
Result is the current channel value modulated with random
(uniform distribution) noise according to the intensity
specified by rvalue. The initial noise intensity
(rvalue=1.0) is the range of one pixel quantum span.
Noise-Uniform
Result is the channel value with uniform noise applied ac-
cording to the intensity specified by rvalue.
As an example, the Assign operator assigns a fixed value to a
channel. For example, this command sets the red channel to the
mid-range value:
gm convert in.bmp -operator red assign "50%" out.bmp
The following applies 50% thresholding to the image and returns
a gray image:
gm convert in.bmp -operator gray threshold "50%" out.bmp
-ordered-dither <channeltype> <NxN>
ordered dither the image
The channel or channels specified in the channeltype argument
are reduced to binary, using an ordered dither method. The
choices for channeltype are All, Intensity, Red, Green, Blue,
Cyan, Magenta, Yellow, Black, and Opacity
When channeltype is "All", the color samples are dithered into a
gray level and then that gray level is stored in the three color
channels. Separately, the opacity channel is dithered into a
bilevel opacity value which is stored in the opacity channel.
When channeltype is "Intensity", only the color samples are
dithered. When channeltype is "opacity" or "matte", only the
opacity channel is dithered. When a color channel is specified,
only that channel is dithered.
The choices for N are 2 through 7. The image is divided into NxN
pixel tiles. In each tile, some or all pixels are turned to
white depending on their intensity. For each N, (N**2)+1 levels
of gray can be represented. For N == 2, 3, or 4, the pixels are
turned to white in an order that maximizes dispersion (i.e., re-
duces granularity), while for N == 5, 6, and 7, they are turned
to white in an order that creates a roughly circular black blob
in the middle of each tile. An attractive "half-tone" looking
image can be obtained by first rotating the image 45 degrees,
performing a 5x5 ordered-dither operation, then rotating it back
to the original orientation and cropping to the original image
dimensions. If the original image is gamma-encoded, it is ad-
viseable to convert it to linear intensity first, e.g., with the
"-gamma 0.45455" option.
-output-directory <directory>
output files to directory
Use -output-directory to specify a directory under which to
write the output files. Normally mogrify overwrites the input
files, but with this option the output files may be written to a
different directory tree so that the input files are preserved.
The algorithm used preserves all of the input path specification
in the output path so that the user-specified input path (in-
cluding any sub-directory part) is appended to the output path.
If the input file lacks an extension, then a suitable extension
is automatically added to the output file. The user is respon-
sible for creating the output directory specified as an argu-
ment, but subdirectories will be created as needed if the -cre-
ate-directories option is supplied. This option may be used to
apply transformations on files from one directory and write the
transformed files to a different directory. In conjunction with
-create-directories, this option is designed to support trans-
forming whole directory trees of files provided that the rela-
tive path of the input file is included as part the list of
filenames.
-orient <orientation>
Set the image orientation attribute
Sets the image orientation attribute. The image orientation at-
tribute is compatible with the TIFF orientation tag (and the
EXIF orientation tag). Accepted values are undefined, TopLeft,
TopRight, BottomRight, BottomLeft, LeftTop, RightTop, RightBot-
tom, LeftBottom, and hyphenated versions thereof (e.g. left-bot-
tom). Please note that GraphicsMagick does not include an EXIF
editor so if an EXIF profile is written to the output image, the
value in the EXIF profile might not match the image. It is pos-
sible for an image file to indicate its orientation in several
different ways simultaneously.
-page <width>x<height>{+-}<x>{+-}<y>{%}{!}{<}{>}
size and location of an image canvas
Use this option to specify the dimensions of the PostScript page
in dots per inch or a TEXT page in pixels. The choices for a
PostScript page are:
11x17 792 1224
Ledger 1224 792
Legal 612 1008
Letter 612 792
LetterSmall 612 792
ArchE 2592 3456
ArchD 1728 2592
ArchC 1296 1728
ArchB 864 1296
ArchA 648 864
A0 2380 3368
A1 1684 2380
A2 1190 1684
A3 842 1190
A4 595 842
A4Small 595 842
A5 421 595
A6 297 421
A7 210 297
A8 148 210
A9 105 148
A10 74 105
B0 2836 4008
B1 2004 2836
B2 1418 2004
B3 1002 1418
B4 709 1002
B5 501 709
C0 2600 3677
C1 1837 2600
C2 1298 1837
C3 918 1298
C4 649 918
C5 459 649
C6 323 459
Flsa 612 936
Flse 612 936
HalfLetter 396 612
For convenience you can specify the page size by media (e.g. A4,
Ledger, etc.). Otherwise, -page behaves much like -geometry
(e.g. -page letter+43+43>).
This option is also used to place subimages when writing to a
multi-image format that supports offsets, such as GIF89 and MNG.
When used for this purpose the offsets are always measured from
the top left corner of the canvas and are not affected by the
-gravity option. To position a GIF or MNG image, use
-page{+-}<x>{+-}<y> (e.g. -page +100+200). When writing to a
MNG file, a -page option appearing ahead of the first image in
the sequence with nonzero width and height defines the width and
height values that are written in the MHDR chunk. Otherwise,
the MNG width and height are computed from the bounding box that
contains all images in the sequence. When writing a GIF89 file,
only the bounding box method is used to determine its dimen-
sions.
For a PostScript page, the image is sized as in -geometry and
positioned relative to the lower left hand corner of the page by
{+-}<xoffset>{+-}<y offset>. Use -page 612x792>, for example, to
center the image within the page. If the image size exceeds the
PostScript page, it is reduced to fit the page. The default
gravity for the -page option is NorthWest, i.e., positive x and
y offset are measured rightward and downward from the top left
corner of the page, unless the -gravity option is present with a
value other than NorthWest.
The default page dimensions for a TEXT image is 612x792.
This option is used in concert with -density.
Use +page to remove the page settings for an image.
-paint <radius>
simulate an oil painting
Each pixel is replaced by the most frequent color in a circular
neighborhood whose width is specified with radius.
-pause <seconds>
pause between animation loops [animate]
Pause for the specified number of seconds before repeating the
animation.
-pause <seconds>
pause between snapshots [import]
Pause for the specified number of seconds before taking the next
snapshot.
-pen <color>
(This option has been replaced by the -fill option)
-ping efficiently determine image characteristics
Use this option to disable reading the image pixels so that im-
age characteristics such as the image dimensions may be obtained
very quickly. For identify, use +ping to force reading the image
pixels so that the pixel read rate may be included in the dis-
played information.
-pointsize <value>
pointsize of the PostScript, X11, or TrueType font
-preview <type>
image preview type
Use this option to affect the preview operation of an image
(e.g. convert file.png -preview Gamma Preview:gamma.png).
Choose from these previews:
Rotate
Shear
Roll
Hue
Saturation
Brightness
Gamma
Spiff
Dull
Grayscale
Quantize
Despeckle
ReduceNoise
AddNoise
Sharpen
Blur
Threshold
EdgeDetect
Spread
Shade
Raise
Segment
Solarize
Swirl
Implode
Wave
OilPaint
CharcoalDrawing
JPEG
The default preview is JPEG.
-process <command>
process a sequence of images using a process module
The command argument has the form module=arg1,arg2,arg3,...,argN
where module is the name of the module to invoke (e.g. "Ana-
lyze") and arg1,arg2,arg3,...,argN are an arbitrary number of
arguments to pass to the process module. The sequence of images
is terminated by the appearance of any option.
If the -process option appears after all of the input images,
all images are processed.
For example:
gm convert logo: -process Analyze= \
-format "%[BrightnessMean],%[BrightnessStddev]" info:-
51952,23294
-profile <filename>
add ICM, IPTC, or generic profile to image
-profile filename adds an ICM (ICC color management), IPTC
(newswire information), or a generic (including Exif) profile to the image
Use +profile icm, +profile iptc, or
+profile profile_name to remove the respective profile.
Multiple profiles may be listed, separated by commas. Profiles may be
excluded from subsequent listed matches by preceding their name with
an exclamation point. For example, +profile '!icm,*' strips
all profiles except for the ICM profile. Use identify
-verbose to find out what profiles are in the image file. Use
+profile "*" to remove all profiles.
Writing the image to a format that does not support profiles will
of course also cause all profiles to be removed. The JPEG and PNG
formats will store any profiles that have been read and not removed.
In JPEG they are stored in APP1 markers, and in PNG they are stored
as hex-coded binary in compressed zTXt chunks, except for the iCC
chunk which is stored in the iCCP chunk.
To extract a profile, the -profile option is not used. Instead,
simply write the file to an image
format such as APP1, 8BIM, ICM, or IPTC.
For example, to extract the Exif data (which is stored in JPEG files
in the APP1 profile), use
gm convert cockatoo.jpg exifdata.app1
Note that GraphicsMagick does not attempt to update any profile
to reflect changes made to the image, e.g., rotation from por-
trait to landscape orientation, so it is possible that the pre-
served profile may contain invalid data.
-preserve-timestamp
preserve the original timestamps of the file
Use this option to preserve the original modification and access
timestamps of the file, even if it has been modified.
+progress
disable progress monitor and busy cursor
By default, when an image is displayed, a progress monitor bar
is shown in the top left corner of an existing image display
window, and the current cursor is replaced with an hourglass
cursor. Use +progress to disable the progress monitor and busy
cursor during display operations. While the progress monitor is
disabled for all operations, the busy cursor continues to be en-
abled for non-display operations such as image processing. This
option is useful for non-interactive display operations, or when
a "clean" look is desired.
-quality <value>
JPEG/MIFF/PNG/TIFF compression level
For the JPEG and MPEG image formats, quality is 0 (lowest image
quality and highest compression) to 100 (best quality but least
effective compression). The default quality is 75. Use the
-sampling-factor option to specify the factors for chroma down-
sampling. To use the same quality value as that found by the
JPEG decoder, use the -define jpeg:preserve-settings flag.
For the MIFF image format, and the TIFF format while using ZIP
compression, quality/10 is the zlib compression level, which is
0 (worst but fastest compression) to 9 (best but slowest). It
has no effect on the image appearance, since the compression is
always lossless.
For the JPEG-2000 image format, quality is mapped using a non-
linear equation to the compression ratio required by the Jasper
library. This non-linear equation is intended to loosely approx-
imate the quality provided by the JPEG v1 format. The default
quality value 75 results in a request for 16:1 compression. The
quality value 100 results in a request for non-lossy compres-
sion.
For the MNG and PNG image formats, the quality value sets the
zlib compression level (quality / 10) and filter-type (quality %
10). Compression levels range from 0 (fastest compression) to
100 (best but slowest). For compression level 0, the Huffman-
only strategy is used, which is fastest but not necessarily the
worst compression.
If filter-type is 4 or less, the specified filter-type is used
for all scanlines:
0: none
1: sub
2: up
3: average
4: Paeth
If filter-type is 5, adaptive filtering is used when quality is
greater than 50 and the image does not have a color map, other-
wise no filtering is used.
If filter-type is 6, adaptive filtering with minimum-sum-of-ab-
solute-values is used.
Only if the output is MNG, if filter-type is 7, the LOCO color
transformation and adaptive filtering with minimum-sum-of-abso-
lute-values are used.
The default is quality is 75, which means nearly the best com-
pression with adaptive filtering. The quality setting has no
effect on the appearance of PNG and MNG images, since the com-
pression is always lossless.
For further information, see the PNG specification.
When writing a JNG image with transparency, two quality values
are required, one for the main image and one for the grayscale
image that conveys the opacity channel. These are written as a
single integer equal to the main image quality plus 1000 times
the opacity quality. For example, if you want to use quality 75
for the main image and quality 90 to compress the opacity data,
use -quality 90075.
For the PNM family of formats (PNM, PGM, and PPM) specify a
quality factor of zero in order to obtain the ASCII variant of
the format. Note that -compress none used to be used to trigger
ASCII output but provided the opposite result of what was ex-
pected as compared with other formats.
For the TIFF format, the JPEG, WebP, Zip, and Zstd compression
algorithms are influenced by the quality value. JPEG and WebP
provide lossy compression so higher quality produces a larger
file with less degradation. The Zip and Zstd compression algo-
rithms (and WebP in lossless mode) are lossless and for these
algorithms a higher ´quality' means to work harder to produce a
smaller file, but with no difference in image quality.
-raise <width>x<height>
lighten or darken image edges
This will create a 3-D effect. See -geometry for details details
about the geometry specification. Offsets are not used.
Use -raise to create a raised effect, otherwise use +raise.
-random-threshold <channeltype> <LOWxHIGH>
random threshold the image
The channel or channels specified in the <channeltype> argument
are reduced to binary, using an random-threshold method. The
choices for channeltype are All, Intensity, Red, Green, Blue,
Cyan, Magenta, Yellow, Black, and Opacity
When channeltype is "All", the color samples are thresholded
into a graylevel and then that gray level is stored in the three
color channels. Separately, the opacity channel is thresholded
into a bilevel opacity value which is stored in the opacity
channel. For each pixel, a new random number is used to estab-
lish the threshold to be used. The threshold never exceeds the
specified maximum (HIGH) and is never less than the specified
minimum (LOW).
When channeltype is "intensity", only the color samples are
thresholded. When channeltype is "opacity" or "matte", only the
opacity channel is thresholded. The other named channels only
threshold the associated channel.
-recolor <matrix>
apply a color translation matrix to image channels
A user supplied color translation matrix (expressed as a text
string) is used to translate/blend the image channels based on
weightings in a supplied matrix which may be of order 3 (color
channels only), 4 (color channels plus opacity), or 5 (color
channels plus opacity and offset). Values in the columns of the
matrix (red, green, blue, opacity) are used as multipliers with
the existing channel values and added together according to the
rows of the matrix. Matrix values are floating point and may be
negative. The offset column (column 5) is purely additive and
is scaled such that 0.0 to 1.0 represents the maximum quantum
range (but values are not limited to this range). The math for
the color translation matrix is similar to that used by Adobe
Flash except that the offset is scaled to 1.0 (divide Flash off-
set by 255 for use with GraphicsMagick) so that the results are
independent of quantum depth.
An identity matrix exists for each matrix order which results in
no change to the image. The translation matrix should be based
on an alteration of the identity matrix.
Identity matrix of order 3
1 0 0
0 1 0
0 0 1
which may be formatted into a convenient matrix argument similar
to (comma is treated as white space):
-recolor "1 0 0, 0 1 0, 0 0 1"
Identity matrix of order 4
1 0 0 0
0 1 0 0
0 0 1 0
0 0 0 1
Identity matrix of order 5. The last row is required to exist
for the purpose of parsing, but is otherwise not used.
1 0 0 0 0
0 1 0 0 0
0 0 1 0 0
0 0 0 1 0
0 0 0 0 1
As an example, an image wrongly in BGR channel order may be con-
verted to RGB using this matrix (blue->red, red->blue):
0 0 1
0 1 0
1 0 0
and an RGB image using standard Rec.709 primaries may be con-
verted to grayscale using this matrix of standard weighting fac-
tors:
0.2126 0.7152 0.0722
0.2126 0.7152 0.0722
0.2126 0.7152 0.0722
and contrast may be reduced by scaling down by 80% and adding a
10% offset:
0.8 0.0 0.0 0.0 0.1
0.0 0.8 0.0 0.0 0.1
0.0 0.0 0.8 0.0 0.1
0.0 0.0 0.0 0.8 0.1
0.0 0.0 0.0 0.0 1.0
-red-primary <x>,<y>
red chromaticity primary point
-region <width>x<height>{+-}<x>{+-}<y>
apply options to a portion of the image
The x and y offsets are treated in the same manner as in -crop.
-remote
perform a X11 remote operation
The -remote command sends a command to a "gm display" or "gm an-
imate" which is already running. The only command recognized at
this time is the name of an image file to load. This capability
is very useful to load new images without needing to restart
GraphicsMagick (e.g. for a slide-show or to use GraphicsMagick
as the display engine for a different GUI). Also see the
+progress option for a way to disable progress indication for a
clean look while loading new images.
-render
render vector operations
Use +render to turn off rendering vector operations. This is
useful when saving the result to vector formats such as MVG or
SVG.
-repage <width>x<height>+xoff+yoff[!]
Adjust image page offsets
Adjust the current image page canvas and position based on a
relative page specification. This option may be used to change
the location of a subframe (e.g. part of an animation) prior to
composition. If the geometry specification is absolute (in-
cludes a '!'), then the offset adjustment is absolute and there
is no adjustment to page width and height, otherwise the page
width and height values are also adjusted based on the current
image dimensions. Use +repage to set the image page offsets to
default.
-resample <horizontal>x<vertical>
Resample image to specified horizontal and vertical resolution
Resize the image so that its rendered size remains the same as
the original at the specified target resolution. Either the cur-
rent image resolution units or the previously set with -units
are used to interpret the argument. For example, if a 300 DPI
image renders at 3 inches by 2 inches on a 300 DPI device, when
the image has been resampled to 72 DPI, it will render at 3
inches by 2 inches on a 72 DPI device. Note that only a small
number of image formats (e.g. JPEG, PNG, and TIFF) are capable
of storing the image resolution. For formats which do not sup-
port an image resolution, the original resolution of the image
must be specified via -density on the command line prior to
specifying the resample resolution.
Note that Photoshop stores and obtains image resolution from a
proprietary embedded profile. If this profile exists in the im-
age, then Photoshop will continue to treat the image using its
former resolution, ignoring the image resolution specified in
the standard file header.
Some image formats (e.g. PNG) require use of metric or english
units so even if the original image used a particular unit sys-
tem, if it is saved to a different format prior to resampling,
then it may be necessary to specify the desired resolution units
using -units since the original units may have been lost. In
other words, do not assume that the resolution units are re-
stored if the image has been saved to a file.
-resize <width>x<height>{%}{@}{!}{<}{>}
resize an image
This is an alias for the -geometry option and it behaves in the
same manner. If the -filter option precedes the -resize option,
the specified filter is used.
There are some exceptions:
When used as a composite option, -resize conveys the preferred
size of the output image, while -geometry conveys the size and
placement of the composite image within the main image.
When used as a montage option, -resize conveys the preferred
size of the montage, while -geometry conveys information about
the tiles.
-roll {+-}<x>{+-}<y>
roll an image vertically or horizontally
See -geometry for details the geometry specification. The x and
y offsets are not affected by the -gravity option.
A negative x offset rolls the image left-to-right. A negative y
offset rolls the image top-to-bottom.
-rotate <degrees>{<}{>}
rotate the image
Positive angles rotate the image in a clockwise direction while
negative angles rotate counter-clockwise.
Use > to rotate the image only if its width exceeds the height.
< rotates the image only if its width is less than the height.
For example, if you specify -rotate "-90>" and the image size is
480x640, the image is not rotated. However, if the image is
640x480, it is rotated by -90 degrees. If you use > or <, en-
close it in quotation marks to prevent it from being misinter-
preted as a file redirection.
Empty triangles left over from rotating the image are filled
with the color defined as background (class backgroundColor).
The color is specified using the format described under the
-fill option.
-sample <geometry>
scale image using pixel sampling
See -geometry for details about the geometry specification.
-sample ignores the -filter selection if the -filter option is
present. Offsets, if present in the geometry string, are ig-
nored, and the -gravity option has no effect.
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
This option specifies the sampling factors to be used by the
DPX, JPEG, MPEG, or YUV encoders for chroma downsampling. The
sampling factor must be specified while reading the raw YUV for-
mat since it is not preserved in the file header. Industry-
standard video subsampling notation such as "4:2:2" may also be
used to specify the sampling factors. "4:2:2" is equivalent to a
specification of "2x1"
The JPEG decoder obtains the original sampling factors (and
quality settings) when a JPEG file is read. To re-use the origi-
nal sampling factors (and quality setting) when JPEG is output,
use the -define jpeg:preserve-settings flag.
-scale <geometry>
scale the image.
See -geometry for details about the geometry specification.
-scale uses a simpler, faster algorithm, and it ignores the
-filter selection if the -filter option is present. Offsets, if
present in the geometry string, are ignored, and the -gravity
option has no effect.
-scene <value>
set scene number
This option sets the scene number of an image or the first image
in an image sequence.
-scenes <value-value>
range of image scene numbers to read
Each image in the range is read with the filename followed by a
period (.) and the decimal scene number. You can change this
behavior by embedding a %d, %0Nd, %o, %0No, %x, or %0Nx printf
format specification in the file name. For example,
gm montage -scenes 5-7 image.miff montage.miff
makes a montage of files image.miff.5, image.miff.6, and im-
age.miff.7, and
gm animate -scenes 0-12 image%02d.miff
animates files image00.miff, image01.miff, through image12.miff.
-screen
specify the screen to capture
This option indicates that the GetImage request used to obtain
the image should be done on the root window, rather than di-
rectly on the specified window. In this way, you can obtain
pieces of other windows that overlap the specified window, and
more importantly, you can capture menus or other popups that are
independent windows but appear over the specified window.
-set <attribute> <value>
set an image attribute
Set a named image attribute. The attribute is set on the cur-
rent (previously specified on command line) image.
+set <attribute>
unset an image attribute
Unset a named image attribute. The attribute is removed from
the current (previously specified on command line) image.
-segment <cluster threshold>x<smoothing threshold>
segment an image
Segment an image by analyzing the histograms of the color compo-
nents and identifying units that are homogeneous with the fuzzy
c-means technique.
Segmentation is a very useful fast and and approximate color
quantization algorithm for scanned printed pages or scanned car-
toons. It may also be used as a special effect. Specify cluster
threshold as the minimum percentage of total pixels in a cluster
before it is considered valid. For huge images containing small
detail, this may need to be a tiny fraction of a percent (e.g.
0.015) so that important detail is not lost. Smoothing thresh-
old eliminates noise in the second derivative of the histogram.
As the value is increased, you can expect a smoother second de-
rivative. The default is 1.5. Add the -verbose option to see a
dump of cluster statistics given the parameters used. The sta-
tistics may be used as a guide to help fine tune the options.
-shade <azimuth>x<elevation>
shade the image using a distant light source
Specify azimuth and elevation as the position of the light
source. Use +shade to return the shading results as a grayscale
image.
-shadow <radius>{x<sigma>}
shadow the montage
-shared-memory
use shared memory
This option specifies whether the utility should attempt to use
shared memory for pixmaps. GraphicsMagick must be compiled with
shared memory support, and the display must support the MIT-SHM
extension. Otherwise, this option is ignored. The default is
True.
-sharpen <radius>{x<sigma>}
sharpen the image
Use a Gaussian operator of the given radius and standard devia-
tion (sigma).
-shave <width>x<height>{%}
shave pixels from the image edges
Specify the width of the region to be removed from both sides of
the image and the height of the regions to be removed from top
and bottom.
-shear <x degrees>x<y degrees>
shear the image along the X or Y axis
Use the specified positive or negative shear angle.
Shearing slides one edge of an image along the X or Y axis, cre-
ating a parallelogram. An X direction shear slides an edge along
the X axis, while a Y direction shear slides an edge along the Y
axis. The amount of the shear is controlled by a shear angle.
For X direction shears, x degrees is measured relative to the Y
axis, and similarly, for Y direction shears y degrees is mea-
sured relative to the X axis.
Empty triangles left over from shearing the image are filled
with the color defined as background (class backgroundColor).
The color is specified using the format described under the
-fill option.
-silent
operate silently
-size <width>x<height>{+offset}
width and height of the image
Use this option to specify the width and height of raw images
whose dimensions are unknown such as GRAY, RGB, or CMYK. In ad-
dition to width and height, use -size with an offset to skip any
header information in the image or tell the number of colors in
a MAP image file, (e.g. -size 640x512+256).
For Photo CD images, choose from these sizes:
192x128
384x256
768x512
1536x1024
3072x2048
Finally, use this option to choose a particular resolution layer
of a JBIG or JPEG image (e.g. -size 1024x768).
-snaps <value>
number of screen snapshots
Use this option to grab more than one image from the X server
screen, to create an animation sequence.
-solarize <factor>
negate all pixels above the threshold level
Specify factor as the percent threshold of the intensity (0 -
99.9%).
This option produces a solarization effect seen when exposing a
photographic film to light during the development process.
-spread <amount>
displace image pixels by a random amount
Amount defines the size of the neighborhood around each pixel to
choose a candidate pixel to swap.
-stegano <offset>
hide watermark within an image
Use an offset to start the image hiding some number of pixels
from the beginning of the image. Note this offset and the image
size. You will need this information to recover the stegano-
graphic image (e.g. display -size 320x256+35 stegano:image.png).
-stereo
composite two images to create a stereo anaglyph
The left side of the stereo pair is saved as the red channel of
the output image. The right side is saved as the green channel.
Red-green stereo glasses are required to properly view the
stereo image.
-strip remove all profiles and text attributes from the image
All embedded profiles and text attributes are stripped from the
image. This is useful for images used for the web, or when out-
put files need to be as small as possible
Be careful not to use this option to remove author, copyright,
and license information that you are required to retain when re-
distributing an image.
-stroke <color>
color to use when stroking a graphic primitive
The color is specified using the format described under the
-fill option.
See -draw for further details.
-strokewidth <value>
set the stroke width
See -draw for further details.
-swirl <degrees>
swirl image pixels about the center
Degrees defines the tightness of the swirl.
-text-font <name>
font for writing fixed-width text
Specifies the name of the preferred font to use in fixed (type-
writer style) formatted text. The default is 14 point Courier.
You can tag a font to specify whether it is a PostScript, True-
Type, or X11 font. For example, Courier.ttf is a TrueType font
and x:fixed is X11.
-texture <filename>
name of texture to tile onto the image background
-threshold <value>{%}
threshold the image
Modify the image such that any pixel sample with an intensity
value greater than the threshold is assigned the maximum inten-
sity (white), or otherwise is assigned the minimum intensity
(black). If a percent prefix is applied, then the threshold is a
percentage of the available range.
To efficiently create a black and white image from a color im-
age, use
gm convert -threshold 50% in.png out.png
The optimum threshold value depends on the nature of the image.
In order to threshold individual channels, use the -operator
subcommand with it's Threshold, Threshold-White, or Threshold-
Black options.
-thumbnail <width>x<height>{%}{@}{!}{<}{>}
resize an image (quickly)
The -thumbnail command resizes the image as quickly as possible,
with more concern for speed than resulting image quality. Re-
gardless, resulting image quality should be acceptable for many
uses. It is primarily intended to be used to generate smaller
versions of the image, but may also be used to enlarge the im-
age. The -thumbnail geometry argument observes the same syntax
and rules as it does for -resize.
-tile <filename>
tile image when filling a graphic primitive
-tile <geometry>
layout of images [montage]
-title <string>
assign title to displayed image [animate, display, montage]
Use this option to assign a specific title to the image. This is
assigned to the image window and is typically displayed in the
window title bar. Optionally you can include the image file-
name, type, width, height, Exif data, or other image attribute
by embedding special format characters described under the -for-
mat option.
For example,
-title "%m:%f %wx%h"
produces an image title of MIFF:bird.miff 512x480 for an image
titled bird.miff and whose width is 512 and height is 480.
-transform
transform the image
This option applies the transformation matrix from a previous
-affine option.
gm convert -affine 2,2,-2,2,0,0 -transform bird.ppm bird.jpg
-transparent <color>
make this color transparent within the image
The color is specified using the format described under the
-fill option.
-treedepth <value>
tree depth for the color reduction algorithm
Normally, this integer value is zero or one. A value of zero or
one causes the use of an optimal tree depth for the color reduc-
tion algorithm
An optimal depth generally allows the best representation of the
source image with the fastest computational speed and the least
amount of memory. However, the default depth is inappropriate
for some images. To assure the best representation, try values
between 2 and 8 for this parameter. Refer to quantize for more
details.
The -colors or -monochrome option, or writing to an image format
which requires color reduction, is required for this option to
take effect.
-trim trim an image
This option removes any edges that are exactly the same color as
the corner pixels. Use -fuzz to make -trim remove edges that
are nearly the same color as the corner pixels.
-type <type>
the image type
Choose from: Bilevel, Grayscale, Palette, PaletteMatte, True-
Color, TrueColorMatte, ColorSeparation, ColorSeparationMatte, or
Optimize.
Normally, when a format supports different subformats such as
bilevel, grayscale, palette, truecolor, and truecolor+alpha, the
encoder will try to choose a suitable subformat based on the na-
ture of the image. The -type option may be used to tailor the
output subformat. By default the output subformat is based on
readily available image information and is usually similar to
the input format.
Specify -type Optimize in order to enable inspecting all pixels
(if necessary) in order to find the most efficient subformat.
Inspecting all of the pixels may be slow for very large images,
particularly if they are stored in a disk cache. If an RGB image
contains only gray pixels, then every pixel in the image must be
inspected in order to decide that the image is actually
grayscale!
Sometimes a specific subformat is desired. For example, to force
a JPEG image to be written in TrueColor RGB format even though
only gray pixels are present, use
gm convert bird.pgm -type TrueColor bird.jpg
Similarly, using -type TrueColorMatte will force the encoder to
write an alpha channel even though the image is opaque, if the
output format supports transparency.
Some pseudo-formats (e.g. the XC format) will respect the re-
quested type if it occurs previously on the command line. For
example, to obtain a DirectClass solid color canvas image rather
than PsuedoClass, use
gm convert -size 640x480 -type TrueColor xc:red red.miff
Likewise, specify -type Bilevel, Grayscale, TrueColor, or True-
ColorMatte prior to reading a Postscript (or PDF file) in order
to influence the type of image that Ghostcript returns. Reading
performance will be dramatically improved for black/white Post-
script if Bilevel is specified, and will be considerably faster
if Grayscale is specified.
-update <seconds>
detect when image file is modified and redisplay.
Suppose that while you are displaying an image the file that is
currently displayed is over-written. display will automatically
detect that the input file has been changed and update the dis-
played image accordingly.
-units <type>
the units of image resolution
Choose from: Undefined, PixelsPerInch, or PixelsPerCentimeter.
This option is normally used in conjunction with the -density
option.
-unsharp <radius>{x<sigma>}{+<amount>}{+<threshold>}
sharpen the image with an unsharp mask operator
The -unsharp option sharpens an image. The image is convolved
with a Gaussian operator of the given radius and standard devia-
tion (sigma). For reasonable results, radius should be larger
than sigma. Use a radius of 0 to have the method select a suit-
able radius.
The parameters are:
radius
The radius of the Gaussian, in pixels, not counting the
center pixel (default 0).
sigma
The standard deviation of the Gaussian, in pixels (default
1.0).
amount
The percentage of the difference between the original and
the blur image that is added back into the original (de-
fault 1.0).
threshold
The threshold, as a fraction of MaxRGB, needed to apply
the difference amount (default 0.05).
-use-pixmap
use the pixmap
-verbose
print detailed information about the image
This information is printed: image scene number; image name; im-
age size; the image class (DirectClass or PseudoClass); the to-
tal number of unique colors; and the number of seconds to read
and transform the image. If the image is DirectClass, the total
number of unique colors is not displayed unless -verbose is
specified twice since it may take quite a long time to compute,
particularly for deep images. If the image is PseudoClass then
its pixels are defined by indexes into a colormap. If the image
is DirectClass then each pixel includes a complete and indepen-
dent color specification.
If -colors is also specified, the total unique colors in the im-
age and color reduction error values are printed. Refer to quan-
tize for a description of these values.
-version
print GraphicsMagick version string
-view <string>
FlashPix viewing parameters
-virtual-pixel <method>
specify contents of "virtual pixels"
This option defines "virtual pixels" for use in operations that
can access pixels outside the boundaries of an image.
Choose from these methods:
Constant
Use the image background color.
Edge
Extend the edge pixel toward infinity (default).
Mirror
Mirror the image.
Tile
Tile the image.
This option affects operations that use virtual pixels such as
-blur, -sharpen, -wave, etc.
-visual <type>
animate images using this X visual type
Choose from these visual classes:
StaticGray
GrayScale
StaticColor
PseudoColor
TrueColor
DirectColor
default
visual id
The X server must support the visual you choose, otherwise an
error occurs. If a visual is not specified, the visual class
that can display the most simultaneous colors on the default
screen is chosen.
-watermark <brightness>x<saturation>
percent brightness and saturation of a watermark
-wave <amplitude>x<wavelength>
alter an image along a sine wave
Specify amplitude and wavelength of the wave.
-white-point <x>,<y>
chromaticity white point
-white-threshold red[,green][,blue][,opacity]
pixels above the threshold become white
Use -white-threshold to set pixels with values above the speci-
fied threshold to maximum value (white). If only one value is
supplied, or the red, green, and blue values are identical, then
intensity thresholding is used. If the color threshold values
are not identical then channel-based thresholding is used, and
color distortion will occur. Specify a negative value (e.g. -1)
if you want a channel to be ignored but you do want to threshold
a channel later in the list. If a percent (%) symbol is ap-
pended, then the values are treated as a percentage of maximum
range.
-window <id>
make image the background of a window
id can be a window id or name. Specify root to select X's root
window as the target window.
By default the image is tiled onto the background of the target
window. If backdrop or -geometry are specified, the image is
surrounded by the background color. Refer to X RESOURCES for
details.
The image will not display on the root window if the image has
more unique colors than the target window colormap allows. Use
-colors to reduce the number of colors.
-window-group
specify the window group
-write <filename>
write an intermediate image [convert, composite]
The current image is written to the specified filename and then
processing continues using that image. The following is an exam-
ple of how several sizes of an image may be generated in one
command (repeat as often as needed):
gm convert input.jpg -resize 50% -write input50.jpg \
-resize 25% input25.jpg
-write <filename>
write the image to a file [display]
If filename already exists, you will be prompted as to whether
it should be overwritten.
By default, the image is written in the format that it was read
in as. To specify a particular image format, prefix filename
with the image type and a colon (e.g., ps:image) or specify the
image type as the filename suffix (e.g., image.ps). Specify file
as - for standard output. If file has the extension .Z or .gz,
the file size is compressed using compress or gzip respectively.
Precede the image file name with | to pipe to a system command.
Use -compress to specify the type of image compression.
The equivalent X resource for this option is writeFilename
(class WriteFilename). See "X Resources", below, for details.
ENVIRONMENT
COLUMNS
Output screen width. Used when formatting text for the screen.
Many Unix systems keep this shell variable up to date, but it
may need to be explicitly exported in order for GraphicsMagick
to see it.
DISPLAY
X11 display ID (host, display number, and screen in the form
hostname:display.screen).
HOME Location of user's home directory. For security reasons, now
only observed by "uninstalled" builds of GraphicsMagick which do
not have their location hard-coded or set by an installer. When
supported, GraphicsMagick searches for configuration files in
$HOME/.magick if the directory exists. See MAGICK_CODER_MOD-
ULE_PATH, MAGICK_CONFIGURE_PATH, and MAGICK_FILTER_MODULE_PATH
if more flexibility is needed.
MAGICK_ACCESS_MONITOR
When set to TRUE, command line monitor mode (enabled by -moni-
tor) will also show files accessed (including temporary files)
and any external commands which are executed. This is useful for
debugging, but also illustrates arguments made available to an
access handler registered by the MagickSetConfirmAccessHandler()
C library function.
MAGICK_CODER_STABILITY
The minimum coder stability level before it will be used. The
available levels are PRIMARY, STABLE, UNSTABLE, and BROKEN. The
default minimum level is UNSTABLE, which means that all avail-
able working coders will be used. The purpose of this option is
to reduce the security exposure (or apparent complexity) due to
the huge number of formats supported. Coders at the PRIMARY
level are commonly used formats with very well maintained imple-
mentations. Coders at the STABLE level are reasonably well main-
tained but represent less used formats. Coders at the UNSTABLE
level either have weak implementations, the file format itself
is weak, or the probability the coder will be needed is vanish-
ingly small. Coders at the BROKEN level are known to often not
work properly or might not be useful in their current state at
all.
MAGICK_CODER_MODULE_PATH
Search path to use when searching for image format coder mod-
ules. This path allows the user to arbitrarily extend the image
formats supported by GraphicsMagick by adding loadable modules
to an arbitrary location rather than copying them into the
GraphicsMagick installation directory. The formatting of the
search path is similar to operating system search paths (i.e.
colon delimited for Unix, and semi-colon delimited for Microsoft
Windows). This user specified search path is used before trying
the default search path.
MAGICK_CONFIGURE_PATH
Search path to use when searching for configuration (.mgk)
files. The formatting of the search path is similar to operat-
ing system search paths (i.e. colon delimited for Unix, and
semi-colon delimited for Microsoft Windows). This user specified
search path is used before trying the default search path.
MAGICK_DEBUG
Debug options (see -debug for details). Setting the configure
debug option via an environment variable (e.g. MAGICK_DEBUG=con-
figure) is necessary to see the complete initialization process,
which includes searching for configuration files.
MAGICK_FILTER_MODULE_PATH
Search path to use when searching for filter process modules
(invoked via -process). This path allows the user to arbitrarily
extend GraphicsMagick's image processing functionality by adding
loadable modules to an arbitrary location rather than copying
them into the GraphicsMagick installation directory. The format-
ting of the search path is similar to operating system search
paths (i.e. colon delimited for Unix, and semi-colon delimited
for Microsoft Windows). This user specified search path is used
before trying the default search path.
MAGICK_GHOSTSCRIPT_PATH
For Microsoft Windows, specify the path to the Ghostscript in-
stallation rather than searching for it via the Windows reg-
istry. This helps in case Ghostscript is not installed via the
Ghostscript Windows installer or the user wants more control
over the Ghostscript used.
MAGICK_HOME
Path to top of GraphicsMagick installation directory. Only ob-
served by "uninstalled" builds of GraphicsMagick which do not
have their location hard-coded or set by an installer.
MAGICK_MMAP_READ
If MAGICK_MMAP_READ is set to TRUE, GraphicsMagick will attempt
to memory-map the input file for reading. This usually substan-
tially improves repeated read performance since the file is al-
ready in memory after the first time it has been read. However,
testing shows that performance may be reduced for files accessed
for the first time since data is accessed via page-faults (upon
first access) and many operating systems fail to do sequential
read-ahead of memory mapped files, and particularly if those
files are accessed over a network. If many large input files
are read, then enabling this option may harm performance by
overloading the operating system's VM system as it then needs to
free unmapped pages and map new ones.
MAGICK_IO_FSYNC
If MAGICK_IO_FSYNC is set to TRUE, then GraphicsMagick will re-
quest that the output file is fully flushed and synchronized to
disk when it is closed. This incurs a performance penalty, but
has the benefit that if the power fails or the system crashes,
the file should be valid on disk. If image files are referenced
from a database, then this option helps assure that the files
referenced by the database are valid.
MAGICK_IOBUF_SIZE
The amount of I/O buffering (in bytes) to use when reading and
writing encoded files. The default is 16384, which is observed
to work well for many cases. The best value for a local filesys-
tem is usually the the native filesystem block size (e.g. 4096,
8192, or even 131,072 for ZFS) in order to minimize the number
of physical disk I/O operations. I/O performance to files ac-
cessed over a network may benefit significantly by tuning this
option. Larger values are not necessarily better (they may be
slower!), and there is rarely any benefit from using values
larger than 32768. Use convert's -verbose option in order to
evaluate read and write rates in pixels per second while keeping
in mind that the operating system will try to cache files in
RAM.
MAGICK_LIMIT_DISK
Maximum amount of disk space allowed for use by the pixel cache.
MAGICK_LIMIT_FILES
Maximum number of open files.
MAGICK_LIMIT_MAP
Maximum size of a memory mapped file allocation. A memory
mapped file consumes memory when the file is accessed, although
the system may reclaim such memory when needed.
MAGICK_LIMIT_MEMORY
Maximum amount of memory to allocate from the heap.
MAGICK_LIMIT_PIXELS
Maximum number of total pixels (image rows times image colums)
to allow for any image which is requested to be created or read.
This is useful to place a limit on how large an image may be.
If the input image file has image dimensions larger than the
pixel limit, then the image memory allocation is denied and an
error is returned immediately. This is a per-image limit and
does not limit the total number of pixels due to multiple image
frames/pages (e.g. multi-page document or an animation).
MAGICK_LIMIT_READ
Maximum number of uncompressed bytes which may be read while de-
coding an image. Each read by the software from the input file
is counted against the total, even if it has been read before.
Decoding fails when the limit is reached. This limit helps de-
fend against highly compressed files (e.g. via gzip), or files
which use complex looping structures, or when data is being read
from a stream (pipe).
MAGICK_LIMIT_WIDTH
Maximum pixel width of an image read, or created.
MAGICK_LIMIT_HEIGHT
Maximum pixel height of an image read, or created.
MAGICK_TMPDIR
Path to directory where GraphicsMagick should write temporary
files. The default is to use the system default, or the location
set by TMPDIR.
TMPDIR For POSIX-compatible systems (Unix-compatible), the path to the
directory where all applications should write temporary files.
Overridden by MAGICK_TMPDIR if it is set.
TMP or TEMP
For Microsoft Windows, the path to the directory where applica-
tions should write temporary files. Overridden by MAGICK_TMPDIR
if it is set.
OMP_NUM_THREADS
As per the OpenMP standard, this specifies the number of threads
to use in parallel regions. Some compilers default the number of
threads to use to the number of processor cores available while
others default to just one thread. See the OpenMP specification
for other standard adjustments and your compiler's manual for
vendor-specific settings.
CONFIGURATION FILES
GraphicsMagick uses a number of XML format configuration files:
colors.mgk
colors configuration file
<?xml version="1.0"?>
<colormap>
<color name="AliceBlue" red="240" green="248" blue="255"
compliance="SVG, X11, XPM" />
</colormap>
delegates.mgk
delegates configuration file
log.mgk
logging configuration file
<?xml version="1.0"?>
<magicklog>
<log events="None" />
<log output="stdout" />
<log filename="Magick-%d.log" />
<log generations="3" />
<log limit="2000" />
<log format="%t %r %u %p %m/%f/%l/%d:\n %e" />
</magicklog>
modules.mgk
loadable modules configuration file
<?xml version="1.0"?>
<modulemap>
<module magick="8BIM" name="META" />
</modulemap>
type.mgk
master type (fonts) configuration file
<?xml version="1.0"?>
<typemap>
<include file="type-windows.mgk" />
<type
name="AvantGarde-Book"
fullname="AvantGarde Book"
family="AvantGarde"
foundry="URW"
weight="400"
style="normal"
stretch="normal"
format="type1"
metrics="/usr/local/share/ghostscript/fonts/a010013l.afm"
glyphs="/usr/local/share/ghostscript/fonts/a010013l.pfb"
/>
</typemap>
GM ANIMATE
Animate displays a sequence of images on any workstation display run-
ning an X server. animate first determines the hardware capabilities of
the workstation. If the number of unique colors in an image is less
than or equal to the number the workstation can support, the image is
displayed in an X window. Otherwise the number of colors in the image
is first reduced to match the color resolution of the workstation be-
fore it is displayed.
This means that a continuous-tone 24 bits-per-pixel image can display
on a 8 bit pseudo-color device or monochrome device. In most instances
the reduced color image closely resembles the original. Alternatively,
a monochrome or pseudo-color image sequence can display on a continu-
ous-tone 24 bits-per-pixel device.
To help prevent color flashing on X server visuals that have colormaps,
animate creates a single colormap from the image sequence. This can be
rather time consuming. You can speed this operation up by reducing the
colors in the image before you "animate" them. Use mogrify to color re-
duce the images to a single colormap. See mogrify(1) for details. Al-
ternatively, you can use a Standard Colormap; or a static, direct, or
true color visual. You can define a Standard Colormap with xstdcmap.
See xstdcmap(1) for details. This method is recommended for colormapped
X server because it eliminates the need to compute a global colormap.
EXAMPLES
To animate a set of images of a cockatoo, use:
gm animate cockatoo.*
To animate a cockatoo image sequence while using the Standard Colormap
best, use:
xstdcmap -best
gm animate -map best cockatoo.*
To animate an image of a cockatoo without a border centered on a back-
drop, use:
gm animate +borderwidth -backdrop cockatoo.*
OPTIONS
For a more detailed description of each option, see Options, above.
-authenticate <string>
decrypt image with this password
-backdrop
display the image centered on a backdrop.
-background <color>
the background color
-bordercolor <color>
the border color
-borderwidth <geometry>
the border width
-chop <width>x<height>{+-}<x>{+-}<y>{%}
remove pixels from the interior of an image
-colormap <type>
define the colormap type
-colors <value>
preferred number of colors in the image
-colorspace <value>
the type of colorspace
-crop <width>x<height>{+-}<x>{+-}<y>{%}
preferred size and location of the cropped image
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-delay <1/100ths of a second>
display the next image after pausing
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-display <host:display[.screen]>
specifies the X server to contact
-dispose <method>
GIF disposal method
-dither
apply Floyd/Steinberg error diffusion to the image
-font <name>
use this font when annotating the image with text
-foreground <color>
define the foreground color
-gamma <value>
level of gamma correction
-geometry <width>x<height>{+-}<x>{+-}<y>{%}{@}{!}{^}{<}{>}
Specify dimension, offset, and resize options.
-help print usage instructions
-iconGeometry <geometry>
specify the icon geometry
-iconic
iconic animation
-interlace <type>
the type of interlacing scheme
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-log <string>
Specify format for debug log
-map <type>
display image using this type.
-matte store matte channel if the image has one
-mattecolor <color>
specify the color to be used with the -frame option
-monitor
show progress indication
-monochrome
transform the image to black and white
-name name an image
-noop NOOP (no option)
-pause <seconds>
pause between animation loops [animate]
-remote
perform a X11 remote operation
-rotate <degrees>{<}{>}
rotate the image
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-scenes <value-value>
range of image scene numbers to read
-shared-memory
use shared memory
-size <width>x<height>{+offset}
width and height of the image
-text-font <name>
font for writing fixed-width text
-title <string>
assign title to displayed image [animate, display, montage]
-treedepth <value>
tree depth for the color reduction algorithm
-trim trim an image
-type <type>
the image type
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
-visual <type>
animate images using this X visual type
-window <id>
make image the background of a window
For a more detailed description of each option, see Options,
above.
Any option you specify on the command line remains in effect for
the group of images following it, until the group is terminated
by the appearance of any option or -noop. For example, to ani-
mate three images, the first with 32 colors, the second with an
unlimited number of colors, and the third with only 16 colors,
use:
gm animate -colors 32 cockatoo.1 -noop cockatoo.2
-colors 16 cockatoo.3
Animate options can appear on the command line or in your X re-
sources file. See X(1). Options on the command line supersede
values specified in your X resources file. Image filenames may
appear in any order on the command line if the image format is
MIFF (refer to miff(5) and the scene keyword is specified in the
image. Otherwise the images will display in the order they ap-
pear on the command line.
MOUSE BUTTONS
Press any button to map or unmap the Command widget. See the next sec-
tion for more information about the Command widget.
COMMAND WIDGET
The Command widget lists a number of sub-menus and commands. They are
Animate
Open
Play
Step
Repeat
Auto Reverse
Speed
Faster
Slower
Direction
Forward
Reverse
Image Info
Help
Quit
Menu items with a indented triangle have a sub-menu. They are repre-
sented above as the indented items. To access a sub-menu item, move the
pointer to the appropriate menu and press a button and drag. When you
find the desired sub-menu item, release the button and the command is
executed. Move the pointer away from the sub-menu if you decide not to
execute a particular command.
KEYBOARD ACCELERATORS
Ctl+O
Press to load an image from a file.
space
Press to display the next image in the sequence.
<
Press to speed-up the display of the images. Refer to
-delay for more information.
>
Press to slow the display of the images. Refer to -delay
for more information.
?
Press to display information about the image. Press any
key or button to erase the information.
This information is printed: image name; image size; and
the total number of unique colors in the image.
F1
Press to display helpful information about animate(1).
Ctl-q
Press to discard all images and exit program.
X RESOURCES
Animate options can appear on the command line or in your X resource
file. Options on the command line supersede values specified in your X
resource file. See X(1) for more information on X resources.
All animate options have a corresponding X resource. In addition, the
animate program uses the following X resources:
background (class Background)
Specifies the preferred color to use for the Image window
background. The default is #ccc.
borderColor (class BorderColor)
Specifies the preferred color to use for the Image window
border. The default is #ccc.
borderWidth (class BorderWidth)
Specifies the width in pixels of the Image window border.
The default is 2.
font (class Font or FontList)
Specifies the name of the preferred font to use in normal
formatted text. The default is 14 point Helvetica.
foreground (class Foreground)
Specifies the preferred color to use for text within the
Image window. The default is black.
geometry (class geometry)
Specifies the preferred size and position of the image
window. It is not necessarily obeyed by all window man-
agers. Offsets, if present, are handled in X(1) style. A
negative x offset is measured from the right edge of the
screen to the right edge of the icon, and a negative y
offset is measured from the bottom edge of the screen to
the bottom edge of the icon.
iconGeometry (class IconGeometry)
Specifies the preferred size and position of the applica-
tion when iconified. It is not necessarily obeyed by all
window managers. Offsets, if present, are handled in the
same manner as in class Geometry.
iconic (class Iconic)
This resource indicates that you would prefer that the ap-
plication's windows initially not be visible as if the
windows had be immediately iconified by you. Window man-
agers may choose not to honor the application's request.
matteColor (class MatteColor)
Specify the color of windows. It is used for the back-
grounds of windows, menus, and notices. A 3D effect is
achieved by using highlight and shadow colors derived from
this color. Default value: #ddd.
name (class Name)
This resource specifies the name under which resources for
the application should be found. This resource is useful
in shell aliases to distinguish between invocations of an
application, without resorting to creating links to alter
the executable file name. The default is the application
name.
sharedMemory (class SharedMemory)
This resource specifies whether animate should attempt use
shared memory for pixmaps. ImageMagick must be compiled
with shared memory support, and the display must support
the MIT-SHM extension. Otherwise, this resource is ig-
nored. The default is True.
text_font (class textFont)
Specifies the name of the preferred font to use in fixed
(typewriter style) formatted text. The default is 14 point
Courier.
title (class Title)
This resource specifies the title to be used for the Image
window. This information is sometimes used by a window
manager to provide some sort of header identifying the
window. The default is the image file name.
GM BATCH
DESCRIPTION
batch executes an arbitary number of the utility commands (e.g. con-
vert) in the form of a simple linear batch script in order to improve
execution efficiency, and/or to allow use as a subordinate co-process
under the control of an arbitrary script or program.
EXAMPLES
To drive 'gm batch' using a shell script (or a program written in any
language), have the script/program send commands to 'gm batch' via its
standard input. Specify that standard input should be used by using
'-' as the file name. The following example converts all files match-
ing '*.jpg' to TIFF format while rotating each file by 90 degrees and
stripping all embedded profiles. The shell script syntax is standard
Unix shell:
for file in *.jpg
do
outfile=`basename $file .jpg`.tiff
echo convert -verbose "'$file'" -rotate 90 \
+profile "'*'" "'$outfile'"
done | gm batch -echo on -feedback on -
We can accomplish the same as the previous example by putting all the
commands in a text file and then specifying the name of the text file
as the script to execute:
for file in *.jpg
do
outfile=`basename $file .jpg`.tiff
echo convert -verbose "'$file'" -rotate 90 \
+profile "'*'" "'$outfile'"
done > script.txt
gm batch -echo on -feedback on script.txt
OPTIONS
Options are processed from left to right and must appear before any
filename argument.
-echo on|off
command echo on or off
Specify on to enable echoing commands to standard output as they
are read or off to disable. The default is off.
-escape unix|windows
Parse using unix or windows syntax
Commands must be parsed from the input stream and escaping needs
to be used to protect spaces or quoting characters in the input.
Specify unix to use unix-style command line parsing or windows
for Microsoft Windows command shell style parsing. The default
depends on if the software is compiled for Microsoft Windows or
for a Unix-type system (including Cygwin on Microsoft Windows).
It is recommended to use unix syntax because it is more powerful
and more portable.
-fail text
text to print if a command fails
When feedback is enabled, this specifies the text to print when
the command fails. The default text is FAIL.
-feedback on|off
enable error feedback
Print text (see -pass and -fail options) feedback after each
command to indicate the result, the default is off.
-help
Prints batch command help.
-pass text
text to print if a command passes
When feedback is enabled, this specifies the text to print when
the command passes. The default text is PASS.
-prompt text
Prompt text to use for command line
If no filename argument was specified, a simple command prompt
appears where you may enter GraphicsMagick commands. The de-
fault prompt is GM>. Use this option to change the prompt to
something else.
-stop-on-error on|off
Specify if command processing stops on error
Normally command processing continues if a command encounters an
error. Specify -stop-on-error on to cause processing to quit
immediately on error.
GM BENCHMARK
DESCRIPTION
benchmark executes an arbitrary gm utility command (e.g. convert) for
one or more loops, and/or a specified execution time, and reports many
execution metrics. For builds using OpenMP, a mode is provided to exe-
cute the benchmark with an increasing number of threads and provide a
report of speedup and multi-thread execution efficiency. If benchmark
is used to execute a command without any additional benchmark options,
then the command is run once.
EXAMPLES
To obtain benchmark information for a single execution of a command:
gm benchmark convert input.ppm -gaussian 0x1 output.ppm
To obtain benchmark information from 100 iterations of the command:
gm benchmark -iterations 100 convert input.ppm \
-gaussian 0x1 output.ppm
To obtain benchmark information by iterating the command until a speci-
fied amount of time (in seconds) has been consumed:
gm benchmark -duration 30 convert input.ppm \
-gaussian 0x1 output.ppm
To obtain a full performance report with an increasing number of
threads (1-32 threads, stepping the number of threads by four each
time):
gm benchmark -duration 3 -stepthreads 4 convert \
input.ppm -gaussian 0x2 output.ppm
Here is the interpretation of the output:
threads - number of threads used.
iter - number of command iterations executed.
user - total user time consumed.
total - total elapsed time consumed.
iter/s - number of command iterations per second.
iter/cpu - amount of CPU time consumed per iteration.
speedup - speedup compared with one thread.
karp-flatt - Karp-Flatt measure of speedup efficiency.
Please note that the reported "speedup" is based on the execution time
of just one thread. A preliminary warm-up pass is used before timing
the first loop in order to ensure that the CPU is brought out of power-
saving modes and that system caches are warmed up. Most modern CPUs
provide a "turbo" mode where the CPU clock speed is increased (e.g. by
a factor of two) when only one or two cores are active. If the CPU
grows excessively hot (due to insufficient cooling), then it may dial
back its clock rates as a form of thermal management. These factors
result in an under-reporting of speedup compared to if "turbo" mode was
disabled and the CPU does not need to worry about thermal management.
The powertop utility available under Linux and Solaris provides a way
to observe CPU core clock rates while a benchmark is running.
OPTIONS
Options are processed from left to right and must appear before any
argument.
-duration duration
duration to run benchmark Specify the number of seconds to run
the benchmark. The command is executed repeatedly until the
specified amount of time has elapsed.
-help
Prints benchmark command help.
-iterations loops
number of command iterations Specify the number of iterations to
run the benchmark. The command is executed repeatedly until the
specified number of iterations has been reached.
-rawcsv
Print results in CSV format Print results in a comma-separated
value (CSV) format which is easy to parse for plotting or im-
porting into a spreadsheet or database. The values reported are
threads, iterations, user_time, and elapsed_time.
-stepthreads step
execute a per-thread benchmark ramp
Execute a per-thread benchmark ramp, incrementing the number of
threads at each step by the specified value. The maximum number
of threads is taken from the standard OMP_NUM_THREADS environ-
ment variable.
GM COMPARE
compare compares two similar images using a specified statistical
method (see -metric) and/or by writing a difference image (-file), with
the altered pixels annotated using a specified method (see -highlight-
style) and color (see -highlight-color). Reference-image is the origi-
nal image and compare-image is the (possibly) altered version, which
should have the same dimensions as reference-image.
EXAMPLES
To compare two images using Mean Square Error (MSE) statistical analy-
sis use:
gm compare -metric mse original.miff compare.miff
To create an annotated difference image use:
gm compare -highlight-style assign -highlight-color purple \
-file diff.miff original.miff compare.miff
OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect only for the image that follows.
All options are reset to their default values after each image is read.
For a more detailed description of each option, see Options, above.
-authenticate <string>
decrypt image with this password
-auto-orient
orient (rotate) image so it is upright
-colorspace <value>
the type of colorspace
-compress <type>
the type of image compression
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-display <host:display[.screen]>
specifies the X server to contact
-endian <type>
specify endianness (MSB, LSB, or Native) of image
-file <filename>
write annotated difference image to file
-help print usage instructions
-highlight-color <color>
pixel annotation color
-highlight-style <style>
pixel annotation style
-interlace <type>
the type of interlacing scheme
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-log <string>
Specify format for debug log
-matte store matte channel if the image has one
-maximum-error <limit>
specifies the maximum amount of total image error
-metric <metric>
comparison metric (MAE, MSE, PAE, PSNR, RMSE)
-monitor
show progress indication
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-size <width>x<height>{+offset}
width and height of the image
-type <type>
the image type
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
For a more detailed description of each option, see Options,
above.
GM COMPOSITE
composite composites (combines) images to create new images. base-im-
age is the base image and change-image contains the changes. ouput-im-
age is the result, and normally has the same dimensions as base-image.
The optional mask-image can be used to provide opacity information for
change-image when it has none or if you want a different mask. A mask
image is typically grayscale and the same size as base-image. If mask-
image is not grayscale, it is converted to grayscale and the resulting
intensities are used as opacity information.
EXAMPLES
To composite an image of a cockatoo with a perch, use:
gm composite cockatoo.miff perch.ras composite.miff
To compute the difference between images in a series, use:
gm composite -compose difference series.2 series.1
difference.miff
To composite an image of a cockatoo with a perch starting at location
(100,150), use:
gm composite -geometry +100+150 cockatoo.miff
perch.ras composite.miff
To tile a logo across your image of a cockatoo, use
gm convert +shade 30x60 cockatoo.miff mask.miff
gm composite -compose bumpmap -tile logo.png
cockatoo.miff mask.miff composite.miff
To composite a red, green, and blue color plane into a single composite
image, try
gm composite -compose CopyGreen green.png red.png
red-green.png
gm composite -compose CopyBlue blue.png red-green.png
gm composite.png
OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect only for the image that follows.
All options are reset to their default values after each image is read.
For a more detailed description of each option, see Options, above.
-authenticate <string>
decrypt image with this password
-background <color>
the background color
-blue-primary <x>,<y>
blue chromaticity primary point
-colors <value>
preferred number of colors in the image
-colorspace <value>
the type of colorspace
-comment <string>
annotate an image with a comment
-compose <operator>
the type of image composition
-compress <type>
the type of image compression
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-displace <horizontal scale>x<vertical scale>
shift image pixels as defined by a displacement map
-display <host:display[.screen]>
specifies the X server to contact
-dispose <method>
GIF disposal method
-dissolve <percent>
dissolve an image into another by the given percent
-dither
apply Floyd/Steinberg error diffusion to the image
-encoding <type>
specify the text encoding
-endian <type>
specify endianness (MSB, LSB, or Native) of image
-filter <type>
use this type of filter when resizing an image
-font <name>
use this font when annotating the image with text
-geometry <width>x<height>{+-}<x>{+-}<y>{%}{@}{!}{^}{<}{>}
Specify dimension, offset, and resize options.
-gravity <type>
direction primitive gravitates to when annotating the image.
-green-primary <x>,<y>
green chromaticity primary point
-help print usage instructions
-interlace <type>
the type of interlacing scheme
-label <name>
assign a label to an image
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-log <string>
Specify format for debug log
-matte store matte channel if the image has one
-monitor
show progress indication
-monochrome
transform the image to black and white
-negate
replace every pixel with its complementary color
-page <width>x<height>{+-}<x>{+-}<y>{%}{!}{<}{>}
size and location of an image canvas
-profile <filename>
add ICM, IPTC, or generic profile to image
-quality <value>
JPEG/MIFF/PNG/TIFF compression level
-recolor <matrix>
apply a color translation matrix to image channels
-red-primary <x>,<y>
red chromaticity primary point
-render
render vector operations
-repage <width>x<height>+xoff+yoff[!]
Adjust image page offsets
-resize <width>x<height>{%}{@}{!}{<}{>}
resize an image
-rotate <degrees>{<}{>}
rotate the image
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-scene <value>
set scene number
-set <attribute> <value>
set an image attribute
+set <attribute>
unset an image attribute
-sharpen <radius>{x<sigma>}
sharpen the image
-size <width>x<height>{+offset}
width and height of the image
-stegano <offset>
hide watermark within an image
-stereo
composite two images to create a stereo anaglyph
-strip remove all profiles and text attributes from the image
-thumbnail <width>x<height>{%}{@}{!}{<}{>}
resize an image (quickly)
-treedepth <value>
tree depth for the color reduction algorithm
-trim trim an image
-type <type>
the image type
-units <type>
the units of image resolution
-unsharp <radius>{x<sigma>}{+<amount>}{+<threshold>}
sharpen the image with an unsharp mask operator
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
-watermark <brightness>x<saturation>
percent brightness and saturation of a watermark
-white-point <x>,<y>
chromaticity white point
-write <filename>
write an intermediate image [convert, composite]
For a more detailed description of each option, see Options,
above.
GM CONJURE
The Magick scripting language (MSL) will primarily benefit those that
want to accomplish custom image processing tasks but do not wish to
program, or those that do not have access to a Perl interpreter or a
compiler. The interpreter is called conjure and here is an example
script:
<?xml version="1.0" encoding="UTF-8"?>
<image size="400x400" >
<read filename="image.gif" />
<get width="base-width" height="base-height" />
<resize geometry="%[dimensions]" />
<get width="width" height="height" />
<print output=
"Image sized from %[base-width]x%[base-height]
to %[width]x%[height].\n" />
<write filename="image.png" />
</image>
invoked with
gm conjure -dimensions 400x400 incantation.msl
All operations will closely follow the key/value pairs defined in Perl-
Magick, unless otherwise noted.
OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect until it is explicitly changed by
specifying the option again with a different effect, or if it is
changed by a statement in the scripting language.
You can define your own keyword/value pairs on the command line. The
script can then use this information when setting values by including
%[keyword] in the string. For example, if you included "-dimensions
400x400" on the command line, as illustrated above, then any string
containing "%[dimensions]" would have 400x400 substituted. The
"%[string]" can be used either an entire string, such as geome-
try="%[dimensions]" or as a part of a string such as filename="%[base-
name].png".
The keyword can be any string except for the following reserved strings
(in any upper, lower, or mixed case variant): debug, help, and verbose,
whose usage is described below.
The value can be any string. If either the keyword or the value con-
tains white space or any symbols that have special meanings to your
shell such as "#", "|", or "%", enclose the string in quotation marks
or use "\" to escape the white space and special symbols.
Keywords and values are case dependent. "Key", "key", and "KEY" would
be three different keywords.
For a more detailed description of each option, see Options, above.
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-help print usage instructions
-log <string>
Specify format for debug log
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
MAGICK SCRIPTING LANGUAGE
The Magick Scripting Language (MSL) presently defines the following el-
ements and their attributes:
<image>
background, color, id, size
Define a new image object. </image> destroys it. Because
of this, if you wish to reference multiple "subimages"
(aka pages or layers), you can embed one image element in-
side of another. For example:
<image>
<read filename="input.png" />
<get width="base-width" height="base-height" />
<image height="base-height" width="base-width">
<image />
<write filename="output.mng" />
</image>
<image size="400x400" />
<group>
Define a new group of image objects. By default, images
are only valid for the life of their <image>element.
<image> -- creates the image
..... -- do stuff with it
</image> -- dispose of the image
However, in a group, all images in that group will stay
around for the life of the group:
<group> -- start a group
<image> -- create an image
.... -- do stuff
</image> -- NOOP
<image> -- create another image
.... -- do more stuff
</image> -- NOOP
<write filename="image.mng" /> -- output
</group> -- dispose of both images
<read>
filename
Read a new image from a disk file.
<read filename="image.gif" />
To read two images use
<read filename="image.gif" />
<read filename="image.png />
<write>
filename
Write the image(s) to disk, either as a single multiple-
image file or multiple ones if necessary.
<write filename=image.tiff" />
<get>
Get any attribute recognized by PerlMagick's GetAt-
tribute() and stores it as an image attribute for later
use. Currently only width and height are supported.
<get width="base-width" height="base-height" />
<print output="Image size is %[base-width]x%[base-height].\n" />
<set>
background, bordercolor, clip-mask, colorspace, density,
magick, mattecolor, opacity. Set an attribute recognized
by PerlMagick's GetAttribute().
<profile>
[profilename]
Read one or more IPTC, ICC or generic profiles from file
and assign to image
<profile iptc="profile.iptc" generic="generic.dat" />
To remove a specified profile use "!" as the filename eg
<profile icm="!" iptc="profile.iptc" />
<border>
fill, geometry, height, width
<blur>
radius, sigma
<charcoal>
radius, sigma
<chop>
geometry, height, width, x, y
<crop>
geometry, height, width, x, y
<composite>
compose, geometry, gravity, image, x, y
<?xml version="1.0" encoding="UTF-8"?>
<group>
<image id="image_01">
<read filename="cloud3.gif"/>
<resize geometry="250x90"/>
</image>
<image id="image_02">
<read filename="cloud4.gif"/>
<resize geometry="190x100"/>
</image>
<image>
<read filename="background.jpg"/>
<composite image="image_01" geometry="+740+470"/>
<composite image="image_02" geometry="+390+415"/>
</image>
<write filename="result.png"/>
</group>
<despeckle>
<emboss>
radius, sigma
<enhance>
<equalize>
<edge>
radius
<flip>
<flop>
<frame>
fill, geometry, height, width, x, y, inner, outer
<flatten>
<get>
height, width
<gamma>
red, green, blue
<image>
background, color, id, size
<implode>
amount
<magnify>
<minify>
<medianfilter>
radius
<normalize>
<oilpaint>
radius
<print>
output
<profile>
[profilename]
<read>
<resize>
blur, filter, geometry, height, width
<roll>
geometry, x, y
<rotate>
degrees
<reducenoise>
radius
<sample>
geometry, height, width
<scale>
geometry, height, width
<sharpen>
radius, sigma
<shave>
geometry, height, width
<shear>
x, y
<solarize>
threshold
<spread>
radius
<stegano>
image
<stereo>
image
<swirl>
degrees
<texture>
image
<threshold>
threshold
<transparent>
color
<trim>
GM CONVERT
Convert converts an input file using one image format to an output file
with a differing image format. In addition, various types of image pro-
cessing can be performed on the converted image during the conversion
process. Convert recognizes the image formats listed in GraphicsMag-
ick(1).
EXAMPLES
To make a thumbnail of a JPEG image, use:
gm convert -size 120x120 cockatoo.jpg -resize 120x120
+profile "*" thumbnail.jpg
In this example, '-size 120x120' gives a hint to the JPEG decoder that
the image is going to be downscaled to 120x120, allowing it to run
faster by avoiding returning full-resolution images to GraphicsMagick
for the subsequent resizing operation. The ´-resize 120x120' specifies
the desired dimensions of the output image. It will be scaled so its
largest dimension is 120 pixels. The ´+profile "*"' removes any ICM,
EXIF, IPTC, or other profiles that might be present in the input and
aren't needed in the thumbnail.
To convert a MIFF image of a cockatoo to a SUN raster image, use:
gm convert cockatoo.miff sun:cockatoo.ras
To convert a multi-page PostScript document to individual FAX pages,
use:
gm convert -monochrome document.ps fax:page
To convert a TIFF image to a PostScript A4 page with the image in the
lower left-hand corner, use:
gm convert -page A4+0+0 image.tiff document.ps
To convert a raw Gray image with a 128 byte header to a portable
graymap, use:
gm convert -depth 8 -size 768x512+128 gray:raw
image.pgm
In this example, "raw" is the input file. Its format is "gray" and it
has the dimensions and number of header bytes specified by the -size
option and the sample depth specified by the -depth option. The output
file is "image.pgm". The suffix ".pgm" specifies its format.
To convert a Photo CD image to a TIFF image, use:
gm convert -size 1536x1024 img0009.pcd image.tiff
gm convert img0009.pcd[4] image.tiff
To create a visual image directory of all your JPEG images, use:
gm convert 'vid:*.jpg' directory.miff
To annotate an image with blue text using font 12x24 at position
(100,100), use:
gm convert -font helvetica -fill blue
-draw "text 100,100 Cockatoo"
bird.jpg bird.miff
To tile a 640x480 image with a JPEG texture with bumps use:
gm convert -size 640x480 tile:bumps.jpg tiled.png
To surround an icon with an ornamental border to use with Mosaic(1),
use:
gm convert -mattecolor "#697B8F" -frame 6x6 bird.jpg
icon.png
To create a MNG animation from a DNA molecule sequence, use:
gm convert -delay 20 dna.* dna.mng
OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect for the set of images that follows,
until the set is terminated by the appearance of any option or -noop.
Some options only affect the decoding of images and others only the en-
coding. The latter can appear after the final group of input images.
For a more detailed description of each option, see Options, above.
-adjoin
join images into a single multi-image file
-affine <matrix>
drawing transform matrix
-antialias
remove pixel aliasing
-append
append a set of images
-asc-cdl <spec>
apply ASC CDL color transform
-authenticate <string>
decrypt image with this password
-auto-orient
orient (rotate) image so it is upright
-average
average a set of images
-background <color>
the background color
-black-threshold red[,green][,blue][,opacity]
pixels below the threshold become black
-blue-primary <x>,<y>
blue chromaticity primary point
-blur <radius>{x<sigma>}
blur the image with a Gaussian operator
-border <width>x<height>
surround the image with a border of color
-bordercolor <color>
the border color
-box <color>
set the color of the annotation bounding box
-channel <type>
the type of channel
-charcoal <factor>
simulate a charcoal drawing
-chop <width>x<height>{+-}<x>{+-}<y>{%}
remove pixels from the interior of an image
-clip apply the clipping path, if one is present
-coalesce
merge a sequence of images
-colorize <value>
colorize the image with the pen color
-colors <value>
preferred number of colors in the image
-colorspace <value>
the type of colorspace
-comment <string>
annotate an image with a comment
-compose <operator>
the type of image composition
-compress <type>
the type of image compression
-contrast
enhance or reduce the image contrast
-convolve <kernel>
convolve image with the specified convolution kernel
-crop <width>x<height>{+-}<x>{+-}<y>{%}
preferred size and location of the cropped image
-cycle <amount>
displace image colormap by amount
-debug <events>
enable debug printout
-deconstruct
break down an image sequence into constituent parts
-define <key>{=<value>},...
add coder/decoder specific options
-delay <1/100ths of a second>
display the next image after pausing
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-despeckle
reduce the speckles within an image
-display <host:display[.screen]>
specifies the X server to contact
-dispose <method>
GIF disposal method
-dither
apply Floyd/Steinberg error diffusion to the image
-draw <string>
annotate an image with one or more graphic primitives
-edge <radius>
detect edges within an image
-emboss <radius>
emboss an image
-encoding <type>
specify the text encoding
-endian <type>
specify endianness (MSB, LSB, or Native) of image
-enhance
apply a digital filter to enhance a noisy image
-equalize
perform histogram equalization to the image
-extent <width>x<height>{+-}<x>{+-}<y>
composite image on background color canvas image
-fill <color>
color to use when filling a graphic primitive
-filter <type>
use this type of filter when resizing an image
-flatten
flatten a sequence of images
-flip create a "mirror image"
-flop create a "mirror image"
-font <name>
use this font when annotating the image with text
-format <string>
output formatted image characteristics
-frame <width>x<height>+<outer bevel width>+<inner bevel width>
surround the image with an ornamental border
-fuzz <distance>{%}
colors within this Euclidean distance are considered equal
-gamma <value>
level of gamma correction
-gaussian <radius>{x<sigma>}
blur the image with a Gaussian operator
-geometry <width>x<height>{+-}<x>{+-}<y>{%}{@}{!}{^}{<}{>}
Specify dimension, offset, and resize options.
-gravity <type>
direction primitive gravitates to when annotating the image.
-green-primary <x>,<y>
green chromaticity primary point
-hald-clut <clut>
apply a Hald CLUT to the image
-help print usage instructions
-implode <factor>
implode image pixels about the center
-intent <type>
use this type of rendering intent when managing the image color
-interlace <type>
the type of interlacing scheme
-label <name>
assign a label to an image
-lat <width>x<height>{+-}<offset>{%}
perform local adaptive thresholding
-level <black_point>{,<gamma>}{,<white_point>}{%}
adjust the level of image contrast
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-list <type>
the type of list
-log <string>
Specify format for debug log
-loop <iterations>
add Netscape loop extension to your GIF animation
-magnify
magnify the image
-map <filename>
choose a particular set of colors from this image
-mask <filename>
Specify a clipping mask
-matte store matte channel if the image has one
-mattecolor <color>
specify the color to be used with the -frame option
-median <radius>
apply a median filter to the image
-minify <factor>
minify the image
-modulate brightness[,saturation[,hue]]
vary the brightness, saturation, and hue of an image
-monitor
show progress indication
-monochrome
transform the image to black and white
-morph <frames>
morphs an image sequence
-mosaic
create a mosaic from an image or an image sequence
-motion-blur <radius>{x<sigma>}{+angle}
Simulate motion blur
-negate
replace every pixel with its complementary color
-noise <radius|type>
add or reduce noise in an image
-noop NOOP (no option)
-normalize
transform image to span the full range of color values
-opaque <color>
change this color to the pen color within the image
-operator channel operator rvalue[%]
apply a mathematical, bitwise, or value operator to an image
channel
-ordered-dither <channeltype> <NxN>
ordered dither the image
-orient <orientation>
Set the image orientation attribute
-page <width>x<height>{+-}<x>{+-}<y>{%}{!}{<}{>}
size and location of an image canvas
-paint <radius>
simulate an oil painting
-pen <color>
(This option has been replaced by the -fill option)
-ping efficiently determine image characteristics
-pointsize <value>
pointsize of the PostScript, X11, or TrueType font
-preview <type>
image preview type
-process <command>
process a sequence of images using a process module
-profile <filename>
add ICM, IPTC, or generic profile to image
-quality <value>
JPEG/MIFF/PNG/TIFF compression level
-raise <width>x<height>
lighten or darken image edges
-random-threshold <channeltype> <LOWxHIGH>
random threshold the image
-recolor <matrix>
apply a color translation matrix to image channels
-red-primary <x>,<y>
red chromaticity primary point
-region <width>x<height>{+-}<x>{+-}<y>
apply options to a portion of the image
-render
render vector operations
-repage <width>x<height>+xoff+yoff[!]
Adjust image page offsets
-resample <horizontal>x<vertical>
Resample image to specified horizontal and vertical resolution
-resize <width>x<height>{%}{@}{!}{<}{>}
resize an image
-roll {+-}<x>{+-}<y>
roll an image vertically or horizontally
-rotate <degrees>{<}{>}
rotate the image
-sample <geometry>
scale image using pixel sampling
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-scale <geometry>
scale the image.
-scene <value>
set scene number
-set <attribute> <value>
set an image attribute
+set <attribute>
unset an image attribute
-segment <cluster threshold>x<smoothing threshold>
segment an image
-shade <azimuth>x<elevation>
shade the image using a distant light source
-sharpen <radius>{x<sigma>}
sharpen the image
-shave <width>x<height>{%}
shave pixels from the image edges
-shear <x degrees>x<y degrees>
shear the image along the X or Y axis
-size <width>x<height>{+offset}
width and height of the image
-solarize <factor>
negate all pixels above the threshold level
-spread <amount>
displace image pixels by a random amount
-strip remove all profiles and text attributes from the image
-stroke <color>
color to use when stroking a graphic primitive
-strokewidth <value>
set the stroke width
-swirl <degrees>
swirl image pixels about the center
-texture <filename>
name of texture to tile onto the image background
-threshold <value>{%}
threshold the image
-thumbnail <width>x<height>{%}{@}{!}{<}{>}
resize an image (quickly)
-tile <filename>
tile image when filling a graphic primitive
-transform
transform the image
-transparent <color>
make this color transparent within the image
-treedepth <value>
tree depth for the color reduction algorithm
-trim trim an image
-type <type>
the image type
-units <type>
the units of image resolution
-unsharp <radius>{x<sigma>}{+<amount>}{+<threshold>}
sharpen the image with an unsharp mask operator
-use-pixmap
use the pixmap
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
-view <string>
FlashPix viewing parameters
-virtual-pixel <method>
specify contents of "virtual pixels"
-wave <amplitude>x<wavelength>
alter an image along a sine wave
-white-point <x>,<y>
chromaticity white point
-white-threshold red[,green][,blue][,opacity]
pixels above the threshold become white
-write <filename>
write an intermediate image [convert, composite]
For a more detailed description of each option, see Options,
above.
GM DISPLAY
Display is a machine architecture independent image processing and dis-
play program. It can display an image on any workstation screen running
an X server. Display can read and write many of the more popular image
formats (e.g. JPEG, TIFF, PNM, Photo CD, etc.).
With display, you can perform these functions on an image:
o load an image from a file
o display the next image
o display the former image
o display a sequence of images as a slide show
o write the image to a file
o print the image to a PostScript printer
o delete the image file
o create a Visual Image Directory
o select the image to display by its thumbnail rather than
name
o undo last image transformation
o copy a region of the image
o paste a region to the image
o restore the image to its original size
o refresh the image
o half the image size
o double the image size
o resize the image
o crop the image
o cut the image
o flop image in the horizontal direction
o flip image in the vertical direction
o rotate the image 90 degrees clockwise
o rotate the image 90 degrees counter-clockwise
o rotate the image
o shear the image
o roll the image
o trim the image edges
o invert the colors of the image
o vary the color brightness
o vary the color saturation
o vary the image hue
o gamma correct the image
o sharpen the image contrast
o dull the image contrast
o perform histogram equalization on the image
o perform histogram normalization on the image
o negate the image colors
o convert the image to grayscale
o set the maximum number of unique colors in the image
o reduce the speckles within an image
o eliminate peak noise from an image
o detect edges within the image
o emboss an image
o segment the image by color
o simulate an oil painting
o simulate a charcoal drawing
o annotate the image with text
o draw on the image
o edit an image pixel color
o edit the image matte information
o composite an image with another
o add a border to the image
o surround image with an ornamental border
o apply image processing techniques to a region of interest
o display information about the image
o zoom a portion of the image
o show a histogram of the image
o display image to background of a window
o set user preferences
o display information about this program
o discard all images and exit program
o change the level of magnification
o display images specified by a World Wide Web (WWW) uniform
resource locator (URL)
EXAMPLES
To scale an image of a cockatoo to exactly 640 pixels in width and 480
pixels in height and position the window at location (200,200), use:
gm display -geometry 640x480+200+200! cockatoo.miff
To display an image of a cockatoo without a border centered on a back-
drop, use:
gm display +borderwidth -backdrop cockatoo.miff
To tile a slate texture onto the root window, use:
gm display -size 1280x1024 -window root slate.png
To display a visual image directory of all your JPEG images, use:
gm display 'vid:*.jpg'
To display a MAP image that is 640 pixels in width and 480 pixels in
height with 256 colors, use:
gm display -size 640x480+256 cockatoo.map
To display an image of a cockatoo specified with a World Wide Web (WWW)
uniform resource locator (URL), use:
gm display ftp://wizards.dupont.com/images/cockatoo.jpg
To display histogram of an image, use:
gm gm convert file.jpg HISTOGRAM:- | gm display -
OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect until it is explicitly changed by
specifying the option again with a different effect. For example to
display three images, the first with 32 colors, the second with an un-
limited number of colors, and the third with only 16 colors, use:
gm display -colors 32 cockatoo.miff -noop duck.miff
-colors 16 macaw.miff
Display options can appear on the command line or in your X resources
file. See X(1). Options on the command line supersede values specified
in your X resources file.
For a more detailed description of each option, see Options, above.
-authenticate <string>
decrypt image with this password
-backdrop
display the image centered on a backdrop.
-background <color>
the background color
-border <width>x<height>
surround the image with a border of color
-bordercolor <color>
the border color
-borderwidth <geometry>
the border width
-colormap <type>
define the colormap type
-colors <value>
preferred number of colors in the image
-colorspace <value>
the type of colorspace
-comment <string>
annotate an image with a comment
-compress <type>
the type of image compression
-contrast
enhance or reduce the image contrast
-crop <width>x<height>{+-}<x>{+-}<y>{%}
preferred size and location of the cropped image
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-delay <1/100ths of a second>
display the next image after pausing
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-despeckle
reduce the speckles within an image
-display <host:display[.screen]>
specifies the X server to contact
-dispose <method>
GIF disposal method
-dither
apply Floyd/Steinberg error diffusion to the image
-edge <radius>
detect edges within an image
-endian <type>
specify endianness (MSB, LSB, or Native) of image
-enhance
apply a digital filter to enhance a noisy image
-filter <type>
use this type of filter when resizing an image
-flip create a "mirror image"
-flop create a "mirror image"
-font <name>
use this font when annotating the image with text
-foreground <color>
define the foreground color
-frame <width>x<height>+<outer bevel width>+<inner bevel width>
surround the image with an ornamental border
-gamma <value>
level of gamma correction
-geometry <width>x<height>{+-}<x>{+-}<y>{%}{@}{!}{^}{<}{>}
Specify dimension, offset, and resize options.
-help print usage instructions
-iconGeometry <geometry>
specify the icon geometry
-iconic
iconic animation
-immutable
make image immutable
-interlace <type>
the type of interlacing scheme
-label <name>
assign a label to an image
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-log <string>
Specify format for debug log
-magnify <factor>
magnify the image
-map <type>
display image using this type.
-matte store matte channel if the image has one
-mattecolor <color>
specify the color to be used with the -frame option
-monitor
show progress indication
-monochrome
transform the image to black and white
-name name an image
-negate
replace every pixel with its complementary color
-noop NOOP (no option)
-page <width>x<height>{+-}<x>{+-}<y>{%}{!}{<}{>}
size and location of an image canvas
+progress
disable progress monitor and busy cursor
-quality <value>
JPEG/MIFF/PNG/TIFF compression level
-raise <width>x<height>
lighten or darken image edges
-remote
perform a X11 remote operation
-roll {+-}<x>{+-}<y>
roll an image vertically or horizontally
-rotate <degrees>{<}{>}
rotate the image
-sample <geometry>
scale image using pixel sampling
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-scenes <value-value>
range of image scene numbers to read
-set <attribute> <value>
set an image attribute
+set <attribute>
unset an image attribute
-segment <cluster threshold>x<smoothing threshold>
segment an image
-shared-memory
use shared memory
-sharpen <radius>{x<sigma>}
sharpen the image
-size <width>x<height>{+offset}
width and height of the image
-text-font <name>
font for writing fixed-width text
-texture <filename>
name of texture to tile onto the image background
-title <string>
assign title to displayed image [animate, display, montage]
-treedepth <value>
tree depth for the color reduction algorithm
-trim trim an image
-type <type>
the image type
-update <seconds>
detect when image file is modified and redisplay.
-use-pixmap
use the pixmap
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
-visual <type>
animate images using this X visual type
-window <id>
make image the background of a window
-window-group
specify the window group
-write <filename>
write the image to a file [display]
For a more detailed description of each option, see Options,
above.
MOUSE BUTTONS
The effects of each button press is described below. Three buttons are
required. If you have a two button mouse, button 1 and 3 are returned.
Press ALT and button 3 to simulate button 2.
1 Press this button to map or unmap the Command widget . See the
next section for more information about the Command widget.
2 Press and drag to define a region of the image to magnify.
3 Press and drag to choose from a select set of display(1) com-
mands. This button behaves differently if the image being dis-
played is a visual image directory. Choose a particular tile of
the directory and press this button and drag to select a command
from a pop-up menu. Choose from these menu items:
Open
Next
Former
Delete
Update
If you choose Open, the image represented by the tile is dis-
played. To return to the visual image directory, choose Next
from the Command widget (refer to Command Widget). Next and
Former moves to the next or former image respectively. Choose
Delete to delete a particular image tile. Finally, choose Update
to synchronize all the image tiles with their respective images.
See montage and miff for more details.
COMMAND WIDGET
The Command widget lists a number of sub-menus and commands. They are
File
Open...
Next
Former
Select...
Save...
Print...
Delete...
Canvas...
Visual Directory...
Quit
Edit
Undo
Redo
Cut
Copy
Paste
View
Half Size
Original Size
Double Size
Resize...
Apply
Refresh
Restore
Transform
Crop
Chop
Flop
Flip
Rotate Right
Rotate Left
Rotate...
Shear...
Roll...
Trim Edges
Enhance
Hue...
Saturation...
Brightness...
Gamma...
Spiff...
Dull
Equalize
Normalize
Negate
GRAYscale
Quantize...
Effects
Despeckle
Emboss
Reduce Noise
Add Noise
Sharpen...
Blur...
Threshold...
Edge Detect...
Spread...
Shade...
Raise...
Segment...
F/X
Solarize...
Swirl...
Implode...
Wave...
Oil Paint...
Charcoal Draw...
Image Edit
Annotate...
Draw...
Color...
Matte...
Composite...
Add Border...
Add Frame...
Comment...
Launch...
Region of Interest...
Miscellany
Image Info
Zoom Image
Show Preview...
Show Histogram
Show Matte
Background...
Slide Show
Preferences...
Help
Overview
Browse Documentation
About Display
Menu items with a indented triangle have a sub-menu. They are repre-
sented above as the indented items. To access a sub-menu item, move the
pointer to the appropriate menu and press button 1 and drag. When you
find the desired sub-menu item, release the button and the command is
executed. Move the pointer away from the sub-menu if you decide not to
execute a particular command.
KEYBOARD ACCELERATORS
Accelerators are one or two key presses that effect a particular com-
mand. The keyboard accelerators that display understands is:
Ctl+O Press to load an image from a file.
space Press to display the next image.
If the image is a multi-paged document such as a PostScript document,
you can skip ahead several pages by preceding this command with a num-
ber. For example to display the fourth page beyond the current page,
press 4space.
backspace Press to display the former image.
If the image is a multi-paged document such as a PostScript document,
you can skip behind several pages by preceding this command with a num-
ber. For example to display the fourth page preceding the current
page, press 4n.
Ctl-S Press to save the image to a file.
Ctl-P Press to print the image to a
PostScript printer.
Ctl-D Press to delete an image file.
Ctl-N Press to create a blank canvas.
Ctl-Q Press to discard all images and exit program.
Ctl+Z Press to undo last image transformation.
Ctl+R Press to redo last image transformation.
Ctl-X Press to cut a region of
the image.
Ctl-C Press to copy a region of
the image.
Ctl-V Press to paste a region to
the image.
< Press to halve the image size.
. Press to return to the original image size.
> Press to double the image size.
% Press to resize the image to a width and height
you specify.
Cmd-A Press to make any image transformations
permanent.
By default, any image size transformations are
applied to the original image to create the
image displayed on the X server. However, the
transformations are not permanent (i.e. the
original image does not change size only the
X image does). For example, if you press ">"
the X image will appear to double in size, but
the original image will in fact remain the same
size. To force the original image to double in
size, press ">" followed by "Cmd-A".
@ Press to refresh the image window.
C Press to crop the image.
[ Press to chop the image.
H Press to flop image in the horizontal direction.
V Press to flip image in the vertical direction.
/ Press to rotate the image 90 degrees clockwise.
\ Press to rotate the image 90 degrees
counter-clockwise.
* Press to rotate the image
the number of degrees you specify.
S Press to shear the image the number of degrees
you specify.
R Press to roll the image.
T Press to trim the image edges.
Shft-H Press to vary the color hue.
Shft-S Press to vary the color saturation.
Shft-L Press to vary the image brightness.
Shft-G Press to gamma correct the image.
Shft-C Press to spiff up the image contrast.
Shft-Z Press to dull the image contrast.
= Press to perform histogram equalization on
the image.
Shft-N Press to perform histogram normalization on
the image.
Shft-~ Press to negate the colors of the image.
. Press to convert the image colors to gray.
Shft-# Press to set the maximum number of unique
colors in the image.
F2 Press to reduce the speckles in an image.
F2 Press to emboss an image.
F4 Press to eliminate peak noise from an image.
F5 Press to add noise to an image.
F6 Press to sharpen an image.
F7 Press to blur image an image.
F8 Press to threshold the image.
F9 Press to detect edges within an image.
F10 Press to displace pixels by a random amount.
F11 Press to shade the image using a distant light
source.
F12 Press to lighten or darken image edges to
create a 3-D effect.
F13 Press to segment the image by color.
Meta-S Press to swirl image pixels about the center.
Meta-I Press to implode image pixels about the center.
Meta-W Press to alter an image along a sine wave.
Meta-P Press to simulate an oil painting.
Meta-C Press to simulate a charcoal drawing.
Alt-X Press to composite the image
with another.
Alt-A Press to annotate the image with text.
Alt-D Press to draw a line on the image.
Alt-P Press to edit an image pixel color.
Alt-M Press to edit the image matte information.
Alt-X Press to composite the image with another.
Alt-A Press to add a border to the image.
Alt-F Press to add a ornamental frame to the image.
Alt-Shft-! Press to add an image comment.
Ctl-A Press to apply image processing techniques to a
region of interest.
Shft-? Press to display information about the image.
Shft-+ Press to map the zoom image window.
Shft-P Press to preview an image enhancement, effect,
or f/x.
F1 Press to display helpful information about
the "display" utility.
Find Press to browse documentation about
GraphicsMagick.
1-9 Press to change the level of magnification.
Use the arrow keys to move the image one pixel up, down, left, or right
within the magnify window. Be sure to first map the magnify window by
pressing button 2.
Press ALT and one of the arrow keys to trim off one pixel from any side
of the image.
X RESOURCES
Display options can appear on the command line or in your X resource
file. Options on the command line supersede values specified in your X
resource file. See X(1) for more information on X resources.
Most display options have a corresponding X resource. In addition, dis-
play uses the following X resources:
background (class Background)
Specifies the preferred color to use for the Image window back-
ground. The default is #ccc.
borderColor (class BorderColor)
Specifies the preferred color to use for the Image window bor-
der. The default is #ccc.
borderWidth (class BorderWidth)
Specifies the width in pixels of the image window border. The
default is 2.
browseCommand (class browseCommand)
Specifies the name of the preferred browser when displaying
GraphicsMagick documentation. The default is netscape %s.
confirmExit (class ConfirmExit)
Display pops up a dialog box to confirm exiting the program
when exiting the program. Set this resource to False to exit
without a confirmation.
displayGamma (class DisplayGamma)
Specifies the gamma of the X server. You can apply separate
gamma values to the red, green, and blue channels of the image
with a gamma value list delineated with slashes (i.e.
1.7/2.3/1.2). The default is 2.2.
displayWarnings (class DisplayWarnings)
Display pops up a dialog box whenever a warning message occurs.
Set this resource to False to ignore warning messages.
font (class FontList)
Specifies the name of the preferred font to use in normal for-
matted text. The default is 14 point Helvetica.
font[1-9] (class Font[1-9])
Specifies the name of the preferred font to use when annotating
the image window with text. The default fonts are fixed, vari-
able, 5x8, 6x10, 7x13bold, 8x13bold, 9x15bold, 10x20, and 12x24.
foreground (class Foreground)
Specifies the preferred color to use for text within the image
window. The default is black.
gammaCorrect (class gammaCorrect)
This resource, if true, will lighten or darken an image of
known gamma to match the gamma of the display (see resource dis-
playGamma). The default is True.
geometry (class Geometry)
Specifies the preferred size and position of the image window.
It is not necessarily obeyed by all window managers.
Offsets, if present, are handled in X(1) style. A negative x
offset is measured from the right edge of the screen to the
right edge of the icon, and a negative y offset is measured from
the bottom edge of the screen to the bottom edge of the icon.
iconGeometry (class IconGeometry)
Specifies the preferred size and position of the application
when iconified. It is not necessarily obeyed by all window man-
agers.
Offsets, if present, are handled in the same manner as in class
Geometry.
iconic (class Iconic)
This resource indicates that you would prefer that the applica-
tion's windows initially not be visible as if the windows had be
immediately iconified by you. Window managers may choose not to
honor the application's request.
magnify (class Magnify)
specifies an integral factor by which the image should be en-
larged. The default is 3. This value only affects the magnifi-
cation window which is invoked with button number 3 after the
image is displayed.
matteColor (class MatteColor)
Specify the color of windows. It is used for the backgrounds of
windows, menus, and notices. A 3D effect is achieved by using
highlight and shadow colors derived from this color. Default
value: #697B8F.
name (class Name)
This resource specifies the name under which resources for the
application should be found. This resource is useful in shell
aliases to distinguish between invocations of an application,
without resorting to creating links to alter the executable file
name. The default is the application name.
pen[1-9] (class Pen[1-9])
Specifies the color of the preferred font to use when annotat-
ing the image window with text. The default colors are black,
blue, green, cyan, gray, red, magenta, yellow, and white.
printCommand (class PrintCommand)
This command is executed whenever Print is issued. In general,
it is the command to print PostScript to your printer. Default
value: lp -c -s %i.
sharedMemory (class SharedMemory)
This resource specifies whether display should attempt use
shared memory for pixmaps. GraphicsMagick must be compiled with
shared memory support, and the display must support the MIT-SHM
extension. Otherwise, this resource is ignored. The default is
True.
textFont (class textFont)
Specifies the name of the preferred font to use in fixed (type-
writer style) formatted text. The default is 14 point Courier.
title (class Title)
This resource specifies the title to be used for the image win-
dow. This information is sometimes used by a window manager to
provide a header identifying the window. The default is the im-
age file name.
undoCache (class UndoCache)
Specifies, in mega-bytes, the amount of memory in the undo edit
cache. Each time you modify the image it is saved in the undo
edit cache as long as memory is available. You can subsequently
undo one or more of these transformations. The default is 16
Megabytes.
usePixmap (class UsePixmap)
Images are maintained as a XImage by default. Set this resource
to True to utilize a server Pixmap instead. This option is use-
ful if your image exceeds the dimensions of your server screen
and you intend to pan the image. Panning is much faster with
Pixmaps than with a XImage. Pixmaps are considered a precious
resource, use them with discretion.
To set the geometry of the Magnify or Pan or window, use the ge-
ometry resource. For example, to set the Pan window geometry to
256x256, use:
gm display.pan.geometry: 256x256
IMAGE LOADING
To select an image to display, choose Open of the File sub-menu from
the Command widget. A file browser is displayed. To choose a particu-
lar image file, move the pointer to the filename and press any button.
The filename is copied to the text window. Next, press Open or press
the RETURN key. Alternatively, you can type the image file name di-
rectly into the text window. To descend directories, choose a directory
name and press the button twice quickly. A scrollbar allows a large
list of filenames to be moved through the viewing area if it exceeds
the size of the list area.
You can trim the list of file names by using shell globbing characters.
For example, type *.jpg to list only files that end with .jpg.
To select your image from the X server screen instead of from a file,
Choose Grab of the Open widget.
VISUAL IMAGE DIRECTORY
To create a Visual Image Directory, choose Visual Directory of the File
sub-menu from the Command widget . A file browser is displayed. To cre-
ate a Visual Image Directory from all the images in the current direc-
tory, press Directory or press the RETURN key. Alternatively, you can
select a set of image names by using shell globbing characters. For ex-
ample, type *.jpg to include only files that end with .jpg. To descend
directories, choose a directory name and press the button twice
quickly. A scrollbar allows a large list of filenames to be moved
through the viewing area if it exceeds the size of the list area.
After you select a set of files, they are turned into thumbnails and
tiled onto a single image. Now move the pointer to a particular thumb-
nail and press button 3 and drag. Finally, select Open. The image rep-
resented by the thumbnail is displayed at its full size. Choose Next
from the File sub-menu of the Command widget to return to the Visual
Image Directory.
IMAGE CUTTING
Note that cut information for image window is not retained for col-
ormapped X server visuals (e.g. StaticColor, StaticColor, GRAYScale,
PseudoColor). Correct cutting behavior may require a TrueColor or Di-
rectColor visual or a Standard Colormap.
To begin, press choose Cut of the Edit sub-menu from the Command wid-
get. Alternatively, press F3 in the image window.
A small window appears showing the location of the cursor in the image
window. You are now in cut mode. In cut mode, the Command widget has
these options:
Help
Dismiss
To define a cut region, press button 1 and drag. The cut region is de-
fined by a highlighted rectangle that expands or contracts as it fol-
lows the pointer. Once you are satisfied with the cut region, release
the button. You are now in rectify mode. In rectify mode, the Command
widget has these options:
Cut
Help
Dismiss
You can make adjustments by moving the pointer to one of the cut rec-
tangle corners, pressing a button, and dragging. Finally, press Cut to
commit your copy region. To exit without cutting the image, press Dis-
miss.
IMAGE COPYING
To begin, press choose Copy of the Edit sub-menu from the Command wid-
get. Alternatively, press F4 in the image window.
A small window appears showing the location of the cursor in the image
window. You are now in copy mode. In copy mode, the Command widget has
these options:
Help
Dismiss
To define a copy region, press button 1 and drag. The copy region is
defined by a highlighted rectangle that expands or contracts as it fol-
lows the pointer. Once you are satisfied with the copy region, release
the button. You are now in rectify mode. In rectify mode, the Command
widget has these options:
Copy
Help
Dismiss
You can make adjustments by moving the pointer to one of the copy rec-
tangle corners, pressing a button, and dragging. Finally, press Copy to
commit your copy region. To exit without copying the image, press Dis-
miss.
IMAGE PASTING
To begin, press choose Paste of the Edit sub-menu from the Command wid-
get. Alternatively, press F5 in the image window.
A small window appears showing the location of the cursor in the image
window. You are now in Paste mode. To exit immediately, press Dismiss.
In Paste mode, the Command widget has these options:
Operators
over
in
out
atop
xor
plus
minus
add
subtract
difference
multiply
bumpmap
replace
Help
Dismiss
Choose a composite operation from the Operators sub-menu of the Command
widget. How each operator behaves is described below. image window is
the image currently displayed on your X server and image is the image
obtained with the File Browser widget.
over The result is the union of the two image shapes, with image ob-
scuring image window in the region of overlap.
in The result is simply image cut by the shape of image window.
None of the image data of image window is in the result.
out The resulting image is image with the shape of image window cut
out.
atop The result is the same shape as image window, with image ob-
scuring image window where the image shapes overlap. Note this
differs from over because the portion of image outside image
window's shape does not appear in the result.
xor The result is the image data from both image and image window
that is outside the overlap region. The overlap region is blank.
plus The result is just the sum of the image data. Output values are
cropped to the maximum value (no overflow). This operation is
independent of the matte channels.
minus The result of image - image window, with underflow cropped to
zero. The matte channel is ignored (set to opaque, full cover-
age).
add The result of image + image window, with overflow wrapping
around (mod MaxRGB+1).
subtract
The result of image - image window, with underflow wrapping
around (mod MaxRGB+1). The add and subtract operators can be
used to perform reversible transformations.
difference
The result of abs(image - image window). This is useful for
comparing two very similar images.
multiply
The result of image * image window. This is useful for the cre-
ation of drop-shadows.
bumpmap
The result of image window shaded by window.
replace
The resulting image is image window replaced with image. Here
the matte information is ignored.
The image compositor requires a matte, or alpha channel in the
image for some operations. This extra channel usually defines a
mask which represents a sort of a cookie-cutter for the image.
This is the case when matte is 255 (full coverage) for pixels
inside the shape, zero outside, and between zero and 255 on the
boundary. If image does not have a matte channel, it is initial-
ized with 0 for any pixel matching in color to pixel location
(0,0), otherwise 255. See Matte Editing for a method of defining
a matte channel.
Note that matte information for image window is not retained for
colormapped X server visuals (e.g. StaticColor, StaticColor,
GrayScale, PseudoColor). Correct compositing behavior may re-
quire a TrueColor or DirectColor visual or a Standard Colormap.
Choosing a composite operator is optional. The default operator
is replace. However, you must choose a location to composite
your image and press button 1. Press and hold the button before
releasing and an outline of the image will appear to help you
identify your location.
The actual colors of the pasted image is saved. However, the
color that appears in image window may be different. For exam-
ple, on a monochrome screen image window will appear black or
white even though your pasted image may have many colors. If the
image is saved to a file it is written with the correct colors.
To assure the correct colors are saved in the final image, any
PseudoClass image is promoted to DirectClass. To force a Pseu-
doClass image to remain PseudoClass, use -colors.
IMAGE CROPPING
To begin, press choose Crop of the Transform submenu from the Command
widget. Alternatively, press C in the image window.
A small window appears showing the location of the cursor in the image
window. You are now in crop mode. In crop mode, the Command widget has
these options:
Help
Dismiss
To define a cropping region, press button 1 and drag. The cropping re-
gion is defined by a highlighted rectangle that expands or contracts as
it follows the pointer. Once you are satisfied with the cropping re-
gion, release the button. You are now in rectify mode. In rectify mode,
the Command widget has these options:
Crop
Help
Dismiss
You can make adjustments by moving the pointer to one of the cropping
rectangle corners, pressing a button, and dragging. Finally, press Crop
to commit your cropping region. To exit without cropping the image,
press Dismiss.
IMAGE CHOPPING
An image is chopped interactively. There is no command line argument to
chop an image. To begin, choose Chop of the Transform sub-menu from the
Command widget. Alternatively, press [ in the Image window.
You are now in Chop mode. To exit immediately, press Dismiss. In Chop
mode, the Command widget has these options:
Direction
horizontal
vertical
Help
Dismiss
If the you choose the horizontal direction (this is the default), the
area of the image between the two horizontal endpoints of the chop line
is removed. Otherwise, the area of the image between the two vertical
endpoints of the chop line is removed.
Select a location within the image window to begin your chop, press and
hold any button. Next, move the pointer to another location in the im-
age. As you move a line will connect the initial location and the
pointer. When you release the button, the area within the image to chop
is determined by which direction you choose from the Command widget.
To cancel the image chopping, move the pointer back to the starting
point of the line and release the button.
IMAGE ROTATION
Press the / key to rotate the image 90 degrees or \ to rotate -90 de-
grees. To interactively choose the degree of rotation, choose Ro-
tate... of the Transform submenu from the Command Widget. Alterna-
tively, press * in the image window.
A small horizontal line is drawn next to the pointer. You are now in
rotate mode. To exit immediately, press Dismiss. In rotate mode, the
Command widget has these options:
Pixel Color
black
blue
cyan
green
gray
red
magenta
yellow
white
Browser...
Direction
horizontal
vertical
Help
Dismiss
Choose a background color from the Pixel Color sub-menu. Additional
background colors can be specified with the color browser. You can
change the menu colors by setting the X resources pen1 through pen9.
If you choose the color browser and press Grab, you can select the
background color by moving the pointer to the desired color on the
screen and press any button.
Choose a point in the image window and press this button and hold.
Next, move the pointer to another location in the image. As you move a
line connects the initial location and the pointer. When you release
the button, the degree of image rotation is determined by the slope of
the line you just drew. The slope is relative to the direction you
choose from the Direction sub-menu of the Command widget.
To cancel the image rotation, move the pointer back to the starting
point of the line and release the button.
IMAGE ANNOTATION
An image is annotated interactively. There is no command line argument
to annotate an image. To begin, choose Annotate of the Image Edit sub-
menu from the Command widget. Alternatively, press a in the image win-
dow.
A small window appears showing the location of the cursor in the image
window. You are now in annotate mode. To exit immediately, press Dis-
miss. In annotate mode, the Command widget has these options:
Font Name
fixed
variable
5x8
6x10
7x13bold
8x13bold
9x15bold
10x20
12x24
Browser...
Font Color
black
blue
cyan
green
gray
red
magenta
yellow
white
transparent
Browser...
Box Color
black
blue
cyan
green
gray
red
magenta
yellow
white
transparent
Browser...
Rotate Text
-90
-45
-30
0
30
45
90
180
Dialog...
Help
Dismiss
Choose a font name from the Font Name sub-menu. Additional font names
can be specified with the font browser. You can change the menu names
by setting the X resources font1 through font9.
Choose a font color from the Font Color sub-menu. Additional font col-
ors can be specified with the color browser. You can change the menu
colors by setting the X resources pen1 through pen9.
If you select the color browser and press Grab, you can choose the font
color by moving the pointer to the desired color on the screen and
press any button.
If you choose to rotate the text, choose Rotate Text from the menu and
select an angle. Typically you will only want to rotate one line of
text at a time. Depending on the angle you choose, subsequent lines may
end up overwriting each other.
Choosing a font and its color is optional. The default font is fixed
and the default color is black. However, you must choose a location to
begin entering text and press a button. An underscore character will
appear at the location of the pointer. The cursor changes to a pencil
to indicate you are in text mode. To exit immediately, press Dismiss.
In text mode, any key presses will display the character at the loca-
tion of the underscore and advance the underscore cursor. Enter your
text and once completed press Apply to finish your image annotation. To
correct errors press BACK SPACE. To delete an entire line of text,
press DELETE. Any text that exceeds the boundaries of the image window
is automatically continued onto the next line.
The actual color you request for the font is saved in the image. How-
ever, the color that appears in your Image window may be different. For
example, on a monochrome screen the text will appear black or white
even if you choose the color red as the font color. However, the image
saved to a file with -write is written with red lettering. To assure
the correct color text in the final image, any PseudoClass image is
promoted to DirectClass (see miff(5)). To force a PseudoClass image to
remain PseudoClass, use -colors.
IMAGE COMPOSITING
An image composite is created interactively. There is no command line
argument to composite an image. To begin, choose Composite of the Image
Edit from the Command widget. Alternatively, press x in the Image win-
dow.
First a popup window is displayed requesting you to enter an image
name. Press Composite, Grab or type a file name. Press Cancel if you
choose not to create a composite image. When you choose Grab, move the
pointer to the desired window and press any button.
If the Composite image does not have any matte information, you are in-
formed and the file browser is displayed again. Enter the name of a
mask image. The image is typically grayscale and the same size as the
composite image. If the image is not grayscale, it is converted to
grayscale and the resulting intensities are used as matte information.
A small window appears showing the location of the cursor in the image
window. You are now in composite mode. To exit immediately, press Dis-
miss. In composite mode, the Command widget has these options:
Operators
over
in
out
atop
xor
plus
minus
add
subtract
difference
bumpmap
replace
Blend
Displace
Help
Dismiss
Choose a composite operation from the Operators sub-menu of the Command
widget. How each operator behaves is described below. image window is
the image currently displayed on your X server and image is the image
obtained
over The result is the union of the two image shapes, with image ob-
scuring image window in the region of overlap.
in The result is simply image cut by the shape of image window.
None of the image data of image window is in the result.
out The resulting image is image with the shape of image window cut
out.
atop The result is the same shape as image window, with image ob-
scuring image window where the image shapes overlap. Note this
differs from over because the portion of image outside image
window's shape does not appear in the result.
xor The result is the image data from both image and image window
that is outside the overlap region. The overlap region is blank.
plus The result is just the sum of the image data. Output values are
cropped to 255 (no overflow). This operation is independent of
the matte channels.
minus The result of image - image window, with underflow cropped to
zero. The matte channel is ignored (set to 255, full coverage).
add The result of image + image window, with overflow wrapping
around (mod 256).
subtract
The result of image - image window, with underflow wrapping
around (mod 256). The add and subtract operators can be used to
perform reversible transformations.
difference
The result of abs(image - image window). This is useful for
comparing two very similar images.
bumpmap
The result of image window shaded by window.
replace
The resulting image is image window replaced with image. Here
the matte information is ignored.
The image compositor requires a matte, or alpha channel in the
image for some operations. This extra channel usually defines a
mask which represents a sort of a cookie-cutter for the image.
This is the case when matte is 255 (full coverage) for pixels
inside the shape, zero outside, and between zero and 255 on the
boundary. If image does not have a matte channel, it is initial-
ized with 0 for any pixel matching in color to pixel location
(0,0), otherwise 255. See Matte Editing for a method of defining
a matte channel.
If you choose blend, the composite operator becomes over. The
image matte channel percent transparency is initialized to fac-
tor. The image window is initialized to (100-factor). Where
factor is the value you specify in the Dialog widget.
Displace shifts the image pixels as defined by a displacement
map. With this option, image is used as a displacement map.
Black, within the displacement map, is a maximum positive dis-
placement. White is a maximum negative displacement and middle
gray is neutral. The displacement is scaled to determine the
pixel shift. By default, the displacement applies in both the
horizontal and vertical directions. However, if you specify
mask, image is the horizontal X displacement and mask the verti-
cal Y displacement.
Note that matte information for image window is not retained for
colormapped X server visuals (e.g. StaticColor, StaticColor,
GrayScale, PseudoColor). Correct compositing behavior may re-
quire a TrueColor or DirectColor visual or a Standard Colormap.
Choosing a composite operator is optional. The default operator
is replace. However, you must choose a location to composite
your image and press button 1. Press and hold the button before
releasing and an outline of the image will appear to help you
identify your location.
The actual colors of the composite image is saved. However, the
color that appears in image window may be different. For exam-
ple, on a monochrome screen Image window will appear black or
white even though your composited image may have many colors. If
the image is saved to a file it is written with the correct col-
ors. To assure the correct colors are saved in the final image,
any PseudoClass image is promoted to DirectClass (see miff). To
force a PseudoClass image to remain PseudoClass, use -colors.
COLOR EDITING
Changing the the color of a set of pixels is performed interactively.
There is no command line argument to edit a pixel. To begin, choose
Color from the Image Edit submenu of the Command widget. Alterna-
tively, press c in the image window.
A small window appears showing the location of the cursor in the image
window. You are now in color edit mode. To exit immediately, press Dis-
miss. In color edit mode, the Command widget has these options:
Method
point
replace
floodfill
reset
Pixel Color
black
blue
cyan
green
gray
red
magenta
yellow
white
Browser...
Border Color
black
blue
cyan
green
gray
red
magenta
yellow
white
Browser...
Fuzz
0
2
4
8
16
Dialog...
Undo
Help
Dismiss
Choose a color editing method from the Method sub-menu of the Command
widget. The point method recolors any pixel selected with the pointer
unless the button is released. The replace method recolors any pixel
that matches the color of the pixel you select with a button press.
Floodfill recolors any pixel that matches the color of the pixel you
select with a button press and is a neighbor. Whereas filltoborder
changes the matte value of any neighbor pixel that is not the border
color. Finally reset changes the entire image to the designated color.
Next, choose a pixel color from the Pixel Color sub-menu. Additional
pixel colors can be specified with the color browser. You can change
the menu colors by setting the X resources pen1 through pen9.
Now press button 1 to select a pixel within the Image window to change
its color. Additional pixels may be recolored as prescribed by the
method you choose. additional pixels by increasing the Delta value.
If the Magnify widget is mapped, it can be helpful in positioning your
pointer within the image (refer to button 2). Alternatively you can se-
lect a pixel to recolor from within the Magnify widget. Move the
pointer to the Magnify widget and position the pixel with the cursor
control keys. Finally, press a button to recolor the selected pixel (or
pixels).
The actual color you request for the pixels is saved in the image. How-
ever, the color that appears in your Image window may be different. For
example, on a monochrome screen the pixel will appear black or white
even if you choose the color red as the pixel color. However, the image
saved to a file with -write is written with red pixels. To assure the
correct color text in the final image, any PseudoClass image is pro-
moted to DirectClass To force a PseudoClass image to remain Pseudo-
Class, use -colors.
MATTE EDITING
Matte information within an image is useful for some operations such as
image compositing. This extra channel usually defines a mask which rep-
resents a sort of a cookie-cutter for the image. This is the case when
matte is 255 (full coverage) for pixels inside the shape, zero outside,
and between zero and 255 on the boundary.
Setting the matte information in an image is done interactively. There
is no command line argument to edit a pixel. To begin, and choose Matte
of the Image Edit sub-menu from the Command widget.
Alternatively, press m in the image window.
A small window appears showing the location of the cursor in the image
window. You are now in matte edit mode. To exit immediately, press Dis-
miss. In matte edit mode, the Command widget has these options:
Method
point
replace
floodfill
reset
Border Color
black
blue
cyan
green
gray
red
magenta
yellow
white
Browser...
Fuzz
0
2
4
8
16
Dialog...
Matte
Undo
Help
Dismiss
Choose a matte editing method from the Method sub-menu of the Command
widget. The point method changes the matte value of the any pixel se-
lected with the pointer until the button is released. The replace
method changes the matte value of any pixel that matches the color of
the pixel you select with a button press. Floodfill changes the matte
value of any pixel that matches the color of the pixel you select with
a button press and is a neighbor. Whereas filltoborder recolors any
neighbor pixel that is not the border color. Finally reset changes the
entire image to the designated matte value. Choose Matte Value and a
dialog appears requesting a matte value. Enter a value between 0 and
255. This value is assigned as the matte value of the selected pixel or
pixels. Now, press any button to select a pixel within the Image win-
dow to change its matte value. You can change the matte value of addi-
tional pixels by increasing the Delta value. The Delta value is first
added then subtracted from the red, green, and blue of the target
color. Any pixels within the range also have their matte value updated.
If the Magnify widget is mapped, it can be helpful in positioning your
pointer within the image (refer to button 2). Alternatively you can se-
lect a pixel to change the matte value from within the Magnify widget.
Move the pointer to the Magnify widget and position the pixel with the
cursor control keys. Finally, press a button to change the matte value
of the selected pixel (or pixels). Matte information is only valid in
a DirectClass image. Therefore, any PseudoClass image is promoted to
DirectClass. Note that matte information for PseudoClass is not re-
tained for colormapped X server visuals (e.g. StaticColor, StaticColor,
GrayScale, PseudoColor) unless you immediately save your image to a
file (refer to Write). Correct matte editing behavior may require a
TrueColor or DirectColor visual or a Standard Colormap.
IMAGE DRAWING
An image is drawn upon interactively. There is no command line argument
to draw on an image. To begin, choose Draw of the Image Edit sub-menu
from the Command widget. Alternatively, press d in the image window.
The cursor changes to a crosshair to indicate you are in draw mode. To
exit immediately, press Dismiss. In draw mode, the Command widget has
these options:
Primitive
point
line
rectangle
fill rectangle
circle
fill circle
ellipse
fill ellipse
polygon
fill polygon
Color
black
blue
cyan
green
gray
red
magenta
yellow
white
transparent
Browser...
Stipple
Brick
Diagonal
Scales
Vertical
Wavy
Translucent
Opaque
Open...
Width
1
2
4
8
16
Dialog...
Undo
Help
Dismiss
Choose a drawing primitive from the Primitive sub-menu.
Next, choose a color from the Color sub-menu. Additional colors can be
specified with the color browser. You can change the menu colors by
setting the X resources pen1 through pen9. The transparent color up-
dates the image matte channel and is useful for image compositing.
If you choose the color browser and press Grab, you can select the
primitive color by moving the pointer to the desired color on the
screen and press any button. The transparent color updates the image
matte channel and is useful for image compositing.
Choose a stipple, if appropriate, from the Stipple sub-menu. Additional
stipples can be specified with the file browser. Stipples obtained from
the file browser must be on disk in the X11 bitmap format.
Choose a line width, if appropriate, from the Width sub-menu. To choose
a specific width select the Dialog widget.
Choose a point in the image window and press button 1 and hold. Next,
move the pointer to another location in the image. As you move, a line
connects the initial location and the pointer. When you release the
button, the image is updated with the primitive you just drew. For
polygons, the image is updated when you press and release the button
without moving the pointer.
To cancel image drawing, move the pointer back to the starting point of
the line and release the button.
REGION OF INTEREST
To begin, press choose Region of Interest of the Pixel Transform sub-
menu from the Command widget. Alternatively, press R in the image win-
dow.
A small window appears showing the location of the cursor in the image
window. You are now in region of interest mode. In region of interest
mode, the Command widget has these options:
Help
Dismiss
To define a region of interest, press button 1 and drag. The region of
interest is defined by a highlighted rectangle that expands or con-
tracts as it follows the pointer. Once you are satisfied with the re-
gion of interest, release the button. You are now in apply mode. In ap-
ply mode the Command widget has these options:
File
Save...
Print...
Edit
Undo
Redo
Transform
Flip
Flop
Rotate Right
Rotate Left
Enhance
Hue...
Saturation...
Brightness...
Gamma...
Spiff
Dull
Equalize
Normalize
Negate
GRAYscale
Quantize...
Effects
Despeckle
Emboss
Reduce Noise
Add Noise
Sharpen...
Blur...
Threshold...
Edge Detect...
Spread...
Shade...
Raise...
Segment...
F/X
Solarize...
Swirl...
Implode...
Wave...
Oil Paint
Charcoal Draw...
Miscellany
Image Info
Zoom Image
Show Preview...
Show Histogram
Show Matte
Help
Dismiss
You can make adjustments to the region of interest by moving the
pointer to one of the rectangle corners, pressing a button, and drag-
ging. Finally, choose an image processing technique from the Command
widget. You can choose more than one image processing technique to ap-
ply to an area. Alternatively, you can move the region of interest be-
fore applying another image processing technique. To exit, press Dis-
miss.
IMAGE PANNING
When an image exceeds the width or height of the X server screen, dis-
play maps a small panning icon. The rectangle within the panning icon
shows the area that is currently displayed in the the image window. To
pan about the image, press any button and drag the pointer within the
panning icon. The pan rectangle moves with the pointer and the image
window is updated to reflect the location of the rectangle within the
panning icon. When you have selected the area of the image you wish to
view, release the button.
Use the arrow keys to pan the image one pixel up, down, left, or right
within the image window.
The panning icon is withdrawn if the image becomes smaller than the di-
mensions of the X server screen.
USER PREFERENCES
Preferences affect the default behavior of display(1). The preferences
are either true or false and are stored in your home directory as .dis-
playrc:
display image centered on a backdrop"
This backdrop covers the entire workstation screen and is
useful for hiding other X window activity while viewing
the image. The color of the backdrop is specified as the
background color. Refer to X Resources for details.
confirm on program exit"
Ask for a confirmation before exiting the display(1) pro-
gram.
correct image for display gamma"
If the image has a known gamma, the gamma is corrected to
match that of the X server (see the X Resource dis-
playGamma).
display warning messages"
Display any warning messages.
apply Floyd/Steinberg error diffusion to image"
The basic strategy of dithering is to trade intensity res-
olution for spatial resolution by averaging the intensi-
ties of several neighboring pixels. Images which suffer
from severe contouring when reducing colors can be im-
proved with this preference.
use a shared colormap for colormapped X visuals"
This option only applies when the default X server visual
is PseudoColor or GRAYScale. Refer to -visual for more de-
tails. By default, a shared colormap is allocated. The im-
age shares colors with other X clients. Some image colors
could be approximated, therefore your image may look very
different than intended. Otherwise the image colors appear
exactly as they are defined. However, other clients may go
technicolor when the image colormap is installed.
display images as an X server pixmap"
Images are maintained as a XImage by default. Set this re-
source to True to utilize a server Pixmap instead. This
option is useful if your image exceeds the dimensions of
your server screen and you intend to pan the image. Pan-
ning is much faster with Pixmaps than with a XImage.
Pixmaps are considered a precious resource, use them with
discretion.
GM IDENTIFY
Identify describes the format and characteristics of one or more
image files as internally supported by the software. It will
also report if an image is incomplete or corrupt. The informa-
tion displayed includes the scene number, the file name, the
width and height of the image, whether the image is colormapped
or not, the number of colors in the image, the number of bytes
in the image, the format of the image (JPEG, PNM, etc.), and fi-
nally the number of seconds in both user time and elapsed time
it took to read and process the image. If -verbose or +ping are
provided as an option, the pixel read rate is also displayed. An
example line output from identify follows:
images/aquarium.miff 640x480 PseudoClass 256c
308135b MIFF 0.000u 0:01
If -verbose is set, expect additional output including any image
comment:
Image: images/aquarium.miff
class: PseudoClass
colors: 256
signature: eb5dca81dd93ae7e6ffae99a527eb5dca8...
matte: False
geometry: 640x480
depth: 8
bytes: 308135
format: MIFF
comments:
Imported from MTV raster image: aquarium.mtv
For some formats, additional format-specific information about
the file will be written if the -debug coder or -debug all op-
tion is used.
IDENTIFY OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect for the set of images immediately
following, until the set is terminated by the appearance of any option
or -noop.
For a more detailed description of each option, see Options, above.
-authenticate <string>
decrypt image with this password
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-format <string>
output formatted image characteristics
-help print usage instructions
-interlace <type>
the type of interlacing scheme
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-log <string>
Specify format for debug log
-ping efficiently determine image characteristics
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-size <width>x<height>{+offset}
width and height of the image
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
For a more detailed description of each option, see Options,
above.
GM IMPORT
Import reads an image from any visible window on an X server and out-
puts it as an image file. You can capture a single window, the entire
screen, or any rectangular portion of the screen. Use display for re-
display, printing, editing, formatting, archiving, image processing,
etc. of the captured image.
The target window can be specified by id, name, or may be selected by
clicking the mouse in the desired window. If you press a button and
then drag, a rectangle will form which expands and contracts as the
mouse moves. To save the portion of the screen defined by the rectan-
gle, just release the button. The keyboard bell is rung once at the be-
ginning of the screen capture and twice when it completes.
EXAMPLES
To select an X window or an area of the screen with the mouse and save
it in the MIFF image format to a file entitled window.miff, use:
gm import window.miff
To select an X window or an area of the screen with the mouse and save
it in the Encapsulated PostScript format to include in another docu-
ment, use:
gm import figure.eps
To capture the entire X server screen in the JPEG image format in a
file entitled root.jpeg, without using the mouse, use:
gm import -window root root.jpeg
To capture the 512x256 area at the upper right corner of the X server
screen in the PNG image format in a well-compressed file entitled cor-
ner.png, without using the mouse, use:
gm import -window root -crop 512x256-0+0 -quality 90
corner.png
OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect until it is explicitly changed by
specifying the option again with a different effect.
Import options can appear on the command line or in your X resources
file. See X(1). Options on the command line supersede values specified
in your X resources file.
For a more detailed description of each option, see Options, above.
-bordercolor <color>
the border color
-colors <value>
preferred number of colors in the image
-colorspace <value>
the type of colorspace
-comment <string>
annotate an image with a comment
-crop <width>x<height>{+-}<x>{+-}<y>{%}
preferred size and location of the cropped image
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-delay <1/100ths of a second>
display the next image after pausing
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-descend
obtain image by descending window hierarchy
-display <host:display[.screen]>
specifies the X server to contact
-dispose <method>
GIF disposal method
-dither
apply Floyd/Steinberg error diffusion to the image
-encoding <type>
specify the text encoding
-endian <type>
specify endianness (MSB, LSB, or Native) of image
-frame include the X window frame in the imported image
-geometry <width>x<height>{+-}<x>{+-}<y>{%}{@}{!}{^}{<}{>}
Specify dimension, offset, and resize options.
-help print usage instructions
-interlace <type>
the type of interlacing scheme
-label <name>
assign a label to an image
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-log <string>
Specify format for debug log
-monitor
show progress indication
-monochrome
transform the image to black and white
-negate
replace every pixel with its complementary color
-page <width>x<height>{+-}<x>{+-}<y>{%}{!}{<}{>}
size and location of an image canvas
-pause <seconds>
pause between snapshots [import]
-ping efficiently determine image characteristics
-pointsize <value>
pointsize of the PostScript, X11, or TrueType font
-quality <value>
JPEG/MIFF/PNG/TIFF compression level
-resize <width>x<height>{%}{@}{!}{<}{>}
resize an image
-rotate <degrees>{<}{>}
rotate the image
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-scene <value>
set scene number
-screen
specify the screen to capture
-set <attribute> <value>
set an image attribute
+set <attribute>
unset an image attribute
-silent
operate silently
-snaps <value>
number of screen snapshots
-thumbnail <width>x<height>{%}{@}{!}{<}{>}
resize an image (quickly)
-transparent <color>
make this color transparent within the image
-trim trim an image
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
For a more detailed description of each option, see Options,
above.
GM MOGRIFY
Mogrify transforms an image or a sequence of images. These transforms
include image scaling, image rotation, color reduction, and others.
Each transmogrified image overwrites the corresponding original image,
unless an option such as -format causes the output filename to be dif-
ferent from the input filename.
The graphics formats supported by mogrify are listed in GraphicsMag-
ick(1).
EXAMPLES
To convert all the TIFF files in a particular directory to JPEG, use:
gm mogrify -format jpeg *.tiff
To convert a directory full of JPEG images to thumbnails, use:
gm mogrify -size 120x120 *.jpg -resize 120x120 +profile "*"
In this example, '-size 120x120' gives a hint to the JPEG decoder that
the images are going to be downscaled to 120x120, allowing it to run
faster by avoiding returning full-resolution images to GraphicsMagick
for the subsequent resizing operation. The ´-resize 120x120' specifies
the desired dimensions of the output images. It will be scaled so its
largest dimension is 120 pixels. The ´+profile "*"' removes any ICM,
EXIF, IPTC, or other profiles that might be present in the input and
aren't needed in the thumbnails.
To scale an image of a cockatoo to exactly 640 pixels in width and 480
pixels in height, use:
gm mogrify -resize 640x480! cockatoo.miff
OPTIONS
Options are processed in command line order. Any option you specify on
the command line remains in effect for the set of images that follows,
until the set is terminated by the appearance of any option or -noop.
For a more detailed description of each option, see Options, above.
-affine <matrix>
drawing transform matrix
-antialias
remove pixel aliasing
-asc-cdl <spec>
apply ASC CDL color transform
-authenticate <string>
decrypt image with this password
-auto-orient
orient (rotate) image so it is upright
-background <color>
the background color
-black-threshold red[,green][,blue][,opacity]
pixels below the threshold become black
-blue-primary <x>,<y>
blue chromaticity primary point
-blur <radius>{x<sigma>}
blur the image with a Gaussian operator
-border <width>x<height>
surround the image with a border of color
-bordercolor <color>
the border color
-channel <type>
the type of channel
-charcoal <factor>
simulate a charcoal drawing
-colorize <value>
colorize the image with the pen color
-colors <value>
preferred number of colors in the image
-colorspace <value>
the type of colorspace
-comment <string>
annotate an image with a comment
-compose <operator>
the type of image composition
-compress <type>
the type of image compression
-contrast
enhance or reduce the image contrast
-convolve <kernel>
convolve image with the specified convolution kernel
-create-directories
create output directory if required
-crop <width>x<height>{+-}<x>{+-}<y>{%}
preferred size and location of the cropped image
-cycle <amount>
displace image colormap by amount
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-delay <1/100ths of a second>
display the next image after pausing
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-despeckle
reduce the speckles within an image
-display <host:display[.screen]>
specifies the X server to contact
-dispose <method>
GIF disposal method
-dither
apply Floyd/Steinberg error diffusion to the image
-draw <string>
annotate an image with one or more graphic primitives
-edge <radius>
detect edges within an image
-emboss <radius>
emboss an image
-encoding <type>
specify the text encoding
-endian <type>
specify endianness (MSB, LSB, or Native) of image
-enhance
apply a digital filter to enhance a noisy image
-equalize
perform histogram equalization to the image
-extent <width>x<height>{+-}<x>{+-}<y>
composite image on background color canvas image
-fill <color>
color to use when filling a graphic primitive
-filter <type>
use this type of filter when resizing an image
-flip create a "mirror image"
-flop create a "mirror image"
-font <name>
use this font when annotating the image with text
-format <type>
the image format type
-frame <width>x<height>+<outer bevel width>+<inner bevel width>
surround the image with an ornamental border
-fuzz <distance>{%}
colors within this Euclidean distance are considered equal
-gamma <value>
level of gamma correction
-gaussian <radius>{x<sigma>}
blur the image with a Gaussian operator
-geometry <width>x<height>{+-}<x>{+-}<y>{%}{@}{!}{^}{<}{>}
Specify dimension, offset, and resize options.
-gravity <type>
direction primitive gravitates to when annotating the image.
-green-primary <x>,<y>
green chromaticity primary point
-hald-clut <clut>
apply a Hald CLUT to the image
-help print usage instructions
-implode <factor>
implode image pixels about the center
-interlace <type>
the type of interlacing scheme
-label <name>
assign a label to an image
-lat <width>x<height>{+-}<offset>{%}
perform local adaptive thresholding
-level <black_point>{,<gamma>}{,<white_point>}{%}
adjust the level of image contrast
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-linewidth
the line width for subsequent draw operations
-list <type>
the type of list
-log <string>
Specify format for debug log
-loop <iterations>
add Netscape loop extension to your GIF animation
-magnify
magnify the image
-map <filename>
choose a particular set of colors from this image
-mask <filename>
Specify a clipping mask
-matte store matte channel if the image has one
-mattecolor <color>
specify the color to be used with the -frame option
-median <radius>
apply a median filter to the image
-minify <factor>
minify the image
-modulate brightness[,saturation[,hue]]
vary the brightness, saturation, and hue of an image
-monitor
show progress indication
-monochrome
transform the image to black and white
-motion-blur <radius>{x<sigma>}{+angle}
Simulate motion blur
-negate
replace every pixel with its complementary color
-noise <radius|type>
add or reduce noise in an image
-noop NOOP (no option)
-normalize
transform image to span the full range of color values
-opaque <color>
change this color to the pen color within the image
-operator channel operator rvalue[%]
apply a mathematical, bitwise, or value operator to an image
channel
-ordered-dither <channeltype> <NxN>
ordered dither the image
-output-directory <directory>
output files to directory
-orient <orientation>
Set the image orientation attribute
-page <width>x<height>{+-}<x>{+-}<y>{%}{!}{<}{>}
size and location of an image canvas
-paint <radius>
simulate an oil painting
-pen <color>
(This option has been replaced by the -fill option)
-pointsize <value>
pointsize of the PostScript, X11, or TrueType font
-profile <filename>
add ICM, IPTC, or generic profile to image
-preserve-timestamp
preserve the original timestamps of the file
-quality <value>
JPEG/MIFF/PNG/TIFF compression level
-raise <width>x<height>
lighten or darken image edges
-random-threshold <channeltype> <LOWxHIGH>
random threshold the image
-recolor <matrix>
apply a color translation matrix to image channels
-red-primary <x>,<y>
red chromaticity primary point
-region <width>x<height>{+-}<x>{+-}<y>
apply options to a portion of the image
-render
render vector operations
-repage <width>x<height>+xoff+yoff[!]
Adjust image page offsets
-resample <horizontal>x<vertical>
Resample image to specified horizontal and vertical resolution
-resize <width>x<height>{%}{@}{!}{<}{>}
resize an image
-roll {+-}<x>{+-}<y>
roll an image vertically or horizontally
-rotate <degrees>{<}{>}
rotate the image
-sample <geometry>
scale image using pixel sampling
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-scale <geometry>
scale the image.
-scene <value>
set scene number
-set <attribute> <value>
set an image attribute
+set <attribute>
unset an image attribute
-segment <cluster threshold>x<smoothing threshold>
segment an image
-shade <azimuth>x<elevation>
shade the image using a distant light source
-sharpen <radius>{x<sigma>}
sharpen the image
-shave <width>x<height>{%}
shave pixels from the image edges
-shear <x degrees>x<y degrees>
shear the image along the X or Y axis
-size <width>x<height>{+offset}
width and height of the image
-solarize <factor>
negate all pixels above the threshold level
-spread <amount>
displace image pixels by a random amount
-strip remove all profiles and text attributes from the image
-stroke <color>
color to use when stroking a graphic primitive
-strokewidth <value>
set the stroke width
-swirl <degrees>
swirl image pixels about the center
-texture <filename>
name of texture to tile onto the image background
-threshold <value>{%}
threshold the image
-thumbnail <width>x<height>{%}{@}{!}{<}{>}
resize an image (quickly)
-tile <filename>
tile image when filling a graphic primitive
-transform
transform the image
-transparent <color>
make this color transparent within the image
-treedepth <value>
tree depth for the color reduction algorithm
-trim trim an image
-type <type>
the image type
-units <type>
the units of image resolution
-unsharp <radius>{x<sigma>}{+<amount>}{+<threshold>}
sharpen the image with an unsharp mask operator
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
-view <string>
FlashPix viewing parameters
-virtual-pixel <method>
specify contents of "virtual pixels"
-wave <amplitude>x<wavelength>
alter an image along a sine wave
-white-point <x>,<y>
chromaticity white point
-white-threshold red[,green][,blue][,opacity]
pixels above the threshold become white
For a more detailed description of each option, see Options,
above.
GM MONTAGE
montage creates a composite image by combining several separate images.
The images are tiled on the composite image with the name of the image
optionally appearing just below the individual tile.
The composite image is constructed in the following manner. First, each
image specified on the command line, except for the last, is scaled to
fit the maximum tile size. The maximum tile size by default is 120x120.
It can be modified with the -geometry command line argument or X re-
source. See Options for more information on command line arguments. See
X(1) for more information on X resources. Note that the maximum tile
size need not be a square.
Next the composite image is initialized with the color specified by the
-background command line argument or X resource. The width and height
of the composite image is determined by the title specified, the maxi-
mum tile size, the number of tiles per row, the tile border width and
height, the image border width, and the label height. The number of
tiles per row specifies how many images are to appear in each row of
the composite image. The default is to have 5 tiles in each row and 4
tiles in each column of the composite. A specific value is specified
with -tile. The tile border width and height, and the image border
width defaults to the value of the X resource -borderwidth. It can be
changed with the -borderwidth or -geometry command line argument or X
resource. The label height is determined by the font you specify with
the -font command line argument or X resource. If you do not specify a
font, a font is chosen that allows the name of the image to fit the
maximum width of a tiled area. The label colors is determined by the
-background and -fill command line argument or X resource. Note, that
if the background and pen colors are the same, labels will not appear.
Initially, the composite image title is placed at the top if one is
specified (refer to -fill). Next, each image is set onto the composite
image, surrounded by its border color, with its name centered just be-
low it. The individual images are left-justified within the width of
the tiled area. The order of the images is the same as they appear on
the command line unless the images have a scene keyword. If a scene
number is specified in each image, then the images are tiled onto the
composite in the order of their scene number. Finally, the last argu-
ment on the command line is the name assigned to the composite image.
By default, the image is written in the MIFF format and can be viewed
or printed with display(1).
Note, that if the number of tiles exceeds the default number of 20 (5
per row, 4 per column), more than one composite image is created. To
ensure a single image is produced, use -tile to increase the number of
tiles to meet or exceed the number of input images.
Finally, to create one or more empty spaces in the sequence of tiles,
use the "NULL:" image format.
Note, a composite MIFF image displayed to an X server with display be-
haves differently than other images. You can think of the composite as
a visual image directory. Choose a particular tile of the composite and
press a button to display it. See display(1) and miff(5)
EXAMPLES
To create a montage of a cockatoo, a parrot, and a hummingbird and
write it to a file called birds, use:
gm montage cockatoo.miff parrot.miff hummingbird.miff
birds.miff
To tile several bird images so that they are at most 256 pixels in
width and 192 pixels in height, surrounded by a red border, and sepa-
rated by 10 pixels of background color, use:
gm montage -geometry 256x192+10+10 -bordercolor red
birds.* montage.miff
To create an unlabeled parrot image, 640 by 480 pixels, and surrounded
by a border of black, use:
gm montage -geometry 640x480 -bordercolor black
-label "" parrot.miff bird.miff
To create an image of an eagle with a textured background, use:
gm montage -texture bumps.jpg eagle.jpg eagle.png
To join several GIF images together without any extraneous graphics
(e.g. no label, no shadowing, no surrounding tile frame), use:
gm montage +frame +shadow +label -tile 5x1
-geometry 50x50+0+0 *.png joined.png
OPTIONS
Any option you specify on the command line remains in effect for the
group of images following it, until the group is terminated by the ap-
pearance of any option or -noop. For example, to make a montage of
three images, the first with 32 colors, the second with an unlimited
number of colors, and the third with only 16 colors, use:
gm montage -colors 32 cockatoo.1 -noop cockatoo.2
-colors 16 cockatoo.3 cockatoos.miff
For a more detailed description of each option, see Options, above.
-adjoin
join images into a single multi-image file
-affine <matrix>
drawing transform matrix
-authenticate <string>
decrypt image with this password
-background <color>
the background color
-blue-primary <x>,<y>
blue chromaticity primary point
-blur <radius>{x<sigma>}
blur the image with a Gaussian operator
-bordercolor <color>
the border color
-borderwidth <geometry>
the border width
-chop <width>x<height>{+-}<x>{+-}<y>{%}
remove pixels from the interior of an image
-colors <value>
preferred number of colors in the image
-colorspace <value>
the type of colorspace
-comment <string>
annotate an image with a comment
-compose <operator>
the type of image composition
-compress <type>
the type of image compression
-crop <width>x<height>{+-}<x>{+-}<y>{%}
preferred size and location of the cropped image
-debug <events>
enable debug printout
-define <key>{=<value>},...
add coder/decoder specific options
-density <width>x<height>
horizontal and vertical resolution in pixels of the image
-depth <value>
depth of the image
-display <host:display[.screen]>
specifies the X server to contact
-dispose <method>
GIF disposal method
-dither
apply Floyd/Steinberg error diffusion to the image
-draw <string>
annotate an image with one or more graphic primitives
-encoding <type>
specify the text encoding
-endian <type>
specify endianness (MSB, LSB, or Native) of image
-fill <color>
color to use when filling a graphic primitive
-filter <type>
use this type of filter when resizing an image
-font <name>
use this font when annotating the image with text
-frame <width>x<height>+<outer bevel width>+<inner bevel width>
surround the image with an ornamental border
-gamma <value>
level of gamma correction
-geometry <width>x<height>{+-}<x>{+-}<y>{%}{@}{!}{^}{<}{>}
Specify dimension, offset, and resize options.
-gravity <type>
direction primitive gravitates to when annotating the image.
-green-primary <x>,<y>
green chromaticity primary point
-help print usage instructions
-interlace <type>
the type of interlacing scheme
-label <name>
assign a label to an image
-limit <type> <value>
Disk, File, Map, Memory, Pixels, Width, Height, Read, or Threads
resource limit
-log <string>
Specify format for debug log
-matte store matte channel if the image has one
-mattecolor <color>
specify the color to be used with the -frame option
-mode <value>
mode of operation
-monitor
show progress indication
-monochrome
transform the image to black and white
-noop NOOP (no option)
-page <width>x<height>{+-}<x>{+-}<y>{%}{!}{<}{>}
size and location of an image canvas
-pen <color>
(This option has been replaced by the -fill option)
-pointsize <value>
pointsize of the PostScript, X11, or TrueType font
-quality <value>
JPEG/MIFF/PNG/TIFF compression level
-red-primary <x>,<y>
red chromaticity primary point
-render
render vector operations
-repage <width>x<height>+xoff+yoff[!]
Adjust image page offsets
-resize <width>x<height>{%}{@}{!}{<}{>}
resize an image
-rotate <degrees>{<}{>}
rotate the image
-sampling-factor <horizontal_factor>x<vertical_factor>
chroma subsampling factors
-scenes <value-value>
range of image scene numbers to read
-shadow <radius>{x<sigma>}
shadow the montage
-sharpen <radius>{x<sigma>}
sharpen the image
-size <width>x<height>{+offset}
width and height of the image
-strip remove all profiles and text attributes from the image
-stroke <color>
color to use when stroking a graphic primitive
-strokewidth <value>
set the stroke width
-texture <filename>
name of texture to tile onto the image background
-thumbnail <width>x<height>{%}{@}{!}{<}{>}
resize an image (quickly)
-tile <geometry>
layout of images [montage]
-title <string>
assign title to displayed image [animate, display, montage]
-transform
transform the image
-transparent <color>
make this color transparent within the image
-treedepth <value>
tree depth for the color reduction algorithm
-trim trim an image
-type <type>
the image type
-verbose
print detailed information about the image
-version
print GraphicsMagick version string
-white-point <x>,<y>
chromaticity white point
For a more detailed description of each option, see Options,
above.
X RESOURCES
Montage options can appear on the command line or in your X resource
file. Options on the command line supersede values specified in your X
resource file. See X(1) for more information on X resources.
All montage options have a corresponding X resource. In addition, mon-
tage uses the following X resources:
background (class Background)
background color
Specifies the preferred color to use for the composite image
background. The default is #ccc.
borderColor (class BorderColor)
border color
Specifies the preferred color to use for the composite image
border. The default is #ccc.
borderWidth (class BorderWidth)
border width
Specifies the width in pixels of the composite image border. The
default is 2.
font (class Font)
font to use
Specifies the name of the preferred font to use when displaying
text within the composite image. The default is 9x15, fixed, or
5x8 determined by the composite image size.
matteColor (class MatteColor)
color of the frame
Specify the color of an image frame. A 3D effect is achieved by
using highlight and shadow colors derived from this color. The
default value is #697B8F.
pen (class Pen)
text color
Specifies the preferred color to use for text within the compos-
ite image. The default is black.
title (class Title)
composite image title
This resource specifies the title to be placed at the top of the
composite image. The default is not to place a title at the top
of the composite image.
GM TIME
DESCRIPTION
time executes an arbitrary gm utility command (e.g. convert) and re-
ports the user and elapsed time. This provides way to measure command
execution times similar to the Unix ´time' command but in a portable
and consistent way.
EXAMPLES
To obtain time information for the execution of a command:
% gm time convert input.ppm -gaussian 0x2 output.ppm
convert input.ppm -gaussian 0x2 output.ppm 22.60s user 0.00s system 2354% cpu 0.960 total
Here is the interpretation of the above output:
user - the total user time consumed.
system - the total system time consumed.
total - the total elapsed time consumed.
OPTIONS
The time command reqires no options other than the gm command to exe-
cute.
GM VERSION
DESCRIPTION
version displays the software release version, build quantum (pixel
sample) depth, web site URL, copyright notice, enabled features sup-
port, configuration parameters, and final build options used to build
the software. The available information depends on how the software
was configured and the host system.
EXAMPLES
To display the version information:
GraphicsMagick 1.3.37 2021-12-12 Q16 http://www.GraphicsMagick.org/
Copyright (C) 2002-2021 GraphicsMagick Group.
Additional copyrights and licenses apply to this software.
See http://www.GraphicsMagick.org/www/Copyright.html for details.
Feature Support:
Native Thread Safe yes
Large Files (> 32 bit) yes
Large Memory (> 32 bit) yes
BZIP yes
DPS no
FlashPix no
FreeType yes
Ghostscript (Library) no
JBIG yes
JPEG-2000 yes
JPEG yes
Little CMS yes
Loadable Modules no
Solaris mtmalloc no
Google perftools tcmalloc no
OpenMP yes (201511 "4.5")
PNG yes
TIFF yes
TRIO no
Solaris umem no
WebP yes
WMF yes
X11 yes
XML yes
ZLIB yes
Host type: x86_64-unknown-linux-gnu
Configured using the command:
./configure ...
Final Build Parameters:
CC = ...
CFLAGS = ...
CPPFLAGS = ...
CXX = ...
CXXFLAGS = ...
LDFLAGS = ...
LIBS = ...
OPTIONS
The version command does not currently support any options.
GraphicsMagick 2022/05/02 gm(1)
Generated by dwww version 1.15 on Sat Jun 1 21:34:37 CEST 2024.