gm - command-line utility to create, edit, compare, convert, or display images
 

gm animate [ options ... ] file [ [ options ... ] file ... ]

gm batch [ options ... ] [ script ]

gm benchmark [ options ... ] subcommand

gm compare [ options ... ] reference-image [ options ... ] compare-image [ 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 ... ] ] output-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

GraphicsMagick's gm provides a suite of utilities for creating, comparing, converting, editing, and displaying images. All of the utilities are provided as sub-commands of a single gm executable:

animate displays an animation (e.g. a GIF file) on any workstation display running an X server.

batch executes an arbitary number of the utility commands (e.g. convert) 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 execution time and other profiling information such as CPU utilization. Benchmark provides various operating modes including executing the command 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 representation 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 create new images.

conjure interprets and executes scripts in the Magick Scripting Language (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 display facility. It can display an image on any workstation display running 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 outputs 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 optionally appearing just below the individual tile.

time executes a subcommand and reports the user, system, and total execution 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
3FR	r--	Hasselblad Photo RAW
8BIM	rw-	Photoshop resource format
8BIMTEXT	rw-	Photoshop resource text format
8BIMWTEXT	rw-	Photoshop resource wide text format
APP1	rw-	Raw application information
APP1JPEG	rw-	Raw JPEG binary data
ART	r--	PF1: 1st Publisher
ARW	r--	Sony Alpha DSLR RAW
AVI	r--	Audio/Visual Interleaved
AVS	rw+	AVS X image
BIE	rw-	Joint Bi-level Image experts Group
		interchange format
BMP	rw+	Microsoft Windows bitmap image
BMP2	-w-	Microsoft Windows bitmap image v2
BMP3	-w-	Microsoft Windows bitmap image v3
CACHE	---	Magick Persistent Cache image format
CALS	rw-	Continuous Acquisition and Life-cycle
		Support Type 1 image
CAPTION	r--	Caption (requires separate size info)
CIN	rw-	Kodak Cineon Format
CMYK	rw-	Raw cyan, magenta, yellow, and black
		samples (8 or 16 bits, depending on
		the image depth)
CMYKA	rw-	Raw cyan, magenta, yellow, black, and
		matte samples (8 or 16 bits, depending
		on the image depth)
CR2	r--	Canon Photo RAW
CRW	r--	Canon Photo RAW
CUR	r--	Microsoft Cursor Icon
CUT	r--	DR Halo
DCM	r--	Digital Imaging and Communications in
		Medicine image
DCR	r--	Kodak Photo RAW
DCX	rw+	ZSoft IBM PC multi-page Paintbrush
DNG	r--	Adobe Digital Negative
DPS	r--	Display PostScript Interpreter
DPX	rw-	Digital Moving Picture Exchange
EPDF	rw-	Encapsulated Portable Document Format
EPI	rw-	Adobe Encapsulated PostScript
		Interchange format
EPS	rw-	Adobe Encapsulated PostScript
EPS2	-w-	Adobe Level II Encapsulated PostScript
EPS3	-w-	Adobe Level III Encapsulated PostScript
EPSF	rw-	Adobe Encapsulated PostScript
EPSI	rw-	Adobe Encapsulated PostScript
		Interchange format
EPT	rw-	Adobe Encapsulated PostScript with MS-DOS
		TIFF preview
EPT2	rw-	Adobe Level II Encapsulated PostScript
		with MS-DOS TIFF preview
EPT3	rw-	Adobe Level III Encapsulated PostScript
		with MS-DOS TIFF preview
EXIF	rw-	Exif digital camera binary data
FAX	rw+	Group 3 FAX (Not TIFF Group3 FAX!)
FITS	rw-	Flexible Image Transport System
FRACTAL	r--	Plasma fractal image
FPX	rw-	FlashPix Format
GIF	rw+	CompuServe graphics interchange format
GIF87	rw-	CompuServe graphics interchange format
		(version 87a)
GRADIENT	r--	Gradual passing from one shade to
		another
GRAY	rw+	Raw gray samples (8/16/32 bits,
		depending on the image depth)
HISTOGRAM	-w-	Histogram of the image
HRZ	r--	HRZ: Slow scan TV
HTML	-w-	Hypertext Markup Language and a
		client-side image map
ICB	rw+	Truevision Targa image
ICC	rw-	ICC Color Profile
ICM	rw-	ICC Color Profile
ICO	r--	Microsoft icon
ICON	r--	Microsoft icon
IDENTITY	r--	Hald CLUT identity image
IMAGE	r--	GraphicsMagick Embedded Image
INFO	-w+	Image descriptive information and
		statistics
IPTC	rw-	IPTC Newsphoto
IPTCTEXT	rw-	IPTC Newsphoto text format
IPTCWTEXT	rw-	IPTC Newsphoto wide text format
JBG	rw+	Joint Bi-level Image experts Group
		interchange format
JBIG	rw+	Joint Bi-level Image experts Group
		interchange format
JNG	rw-	JPEG Network Graphics
JP2	rw-	JPEG-2000 JP2 File Format Syntax
JPC	rw-	JPEG-2000 Code Stream Syntax
JPEG	rw-	Joint Photographic Experts Group
		JFIF format
JPG	rw-	Joint Photographic Experts Group
		JFIF format
K25	r--	Kodak Photo RAW
KDC	r--	Kodak Photo RAW
LABEL	r--	Text image format
M2V	rw+	MPEG-2 Video Stream
MAP	rw-	Colormap intensities and indices
MAT	r--	MATLAB image format
MATTE	-w+	MATTE format
MIFF	rw+	Magick Image File Format
MNG	rw+	Multiple-image Network Graphics
MONO	rw-	Bi-level bitmap in least-significant-
		-byte-first order
MPC	rw+	Magick Persistent Cache image format
MPEG	rw+	MPEG-1 Video Stream
MPG	rw+	MPEG-1 Video Stream
MRW	r--	Minolta Photo Raw
MSL	r--	Magick Scripting Language
MTV	rw+	MTV Raytracing image format
MVG	rw-	Magick Vector Graphics
NEF	r--	Nikon Electronic Format
NULL	r--	Constant image of uniform color
OTB	rw-	On-the-air bitmap
P7	rw+	Xv thumbnail format
PAL	rw-	16bit/pixel interleaved YUV
PALM	rw-	Palm Pixmap
PBM	rw+	Portable bitmap format (black and white)
PCD	rw-	Photo CD
PCDS	rw-	Photo CD
PCL	-w-	Page Control Language
PCT	rw-	Apple Macintosh QuickDraw/PICT
PCX	rw-	ZSoft IBM PC Paintbrush
PDB	rw+	Palm Database ImageViewer Format
PDF	rw+	Portable Document Format
PEF	r--	Pentax Electronic File
PFA	r--	TrueType font
PFB	r--	TrueType font
PGM	rw+	Portable graymap format (gray scale)
PGX	r--	JPEG-2000 VM Format
PICON	rw-	Personal Icon
PICT	rw-	Apple Macintosh QuickDraw/PICT
PIX	r--	Alias/Wavefront RLE image format
PLASMA	r--	Plasma fractal image
PNG	rw-	Portable Network Graphics
PNG24	rw-	Portable Network Graphics, 24 bit RGB
		opaque only
PNG32	rw-	Portable Network Graphics, 32 bit RGBA
		semitransparency OK
PNG8	rw-	Portable Network Graphics, 8-bit
		indexed, binary transparency only
PNM	rw+	Portable anymap
PPM	rw+	Portable pixmap format (color)
PREVIEW	-w-	Show a preview an image enhancement,
		effect, or f/x
PS	rw+	Adobe PostScript
PS2	-w+	Adobe Level II PostScript
PS3	-w+	Adobe Level III PostScript
PSD	rw-	Adobe Photoshop bitmap
PTIF	rw-	Pyramid encoded TIFF
PWP	r--	Seattle Film Works
RAF	r--	Fuji Photo RAW
RAS	rw+	SUN Rasterfile
RGB	rw+	Raw red, green, and blue samples
RGBA	rw+	Raw red, green, blue, and matte samples
RLA	r--	Alias/Wavefront image
RLE	r--	Utah Run length encoded image
SCT	r--	Scitex HandShake
SFW	r--	Seattle Film Works
SGI	rw+	Irix RGB image
SHTML	-w-	Hypertext Markup Language and a
		client-side image map
STEGANO	r--	Steganographic image
SUN	rw+	SUN Rasterfile
SVG	rw+	Scalable Vector Gaphics
TEXT	rw+	Raw text
TGA	rw+	Truevision Targa image
TIFF	rw+	Tagged Image File Format
TILE	r--	Tile image with a texture
TIM	r--	PSX TIM
TOPOL	r--	TOPOL X Image
TTF	r--	TrueType font
TXT	rw+	Raw text
UIL	-w-	X-Motif UIL table
UYVY	rw-	16bit/pixel interleaved YUV
VDA	rw+	Truevision Targa image
VICAR	rw-	VICAR rasterfile format
VID	rw+	Visual Image Directory
VIFF	rw+	Khoros Visualization image
VST	rw+	Truevision Targa image
WBMP	rw-	Wireless Bitmap (level 0) image
WMF	r--	Windows Metafile
WPG	r--	Word Perfect Graphics
X	rw-	X Image
X3F	r--	Foveon X3 (Sigma/Polaroid) RAW
XBM	rw-	X Windows system bitmap (black
		and white)
XC	r--	Constant image uniform color
XCF	r--	GIMP image
XMP	rw-	Adobe XML metadata
XPM	rw-	X Windows system pixmap (color)
XV	rw+	Khoros Visualization image
XWD	rw-	X Windows system window dump (color)
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 libraries. 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 +adjoin 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 compiled 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, respectively, 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.

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:image) or specify the image type as the filename suffix (i.e.image.ps). The magic number takes precedence over the filename suffix and the prefix takes precedence over the magic number and the suffix in input files. When a file is read, its magic number is stored in the "image->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 meaning. 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 standard 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.

Finally, when running on platforms that allow it, precede the image file name with | to pipe to or from a system command (this feature is not available on VMS, Win32 and Macintosh platforms). Use a backslash or quotation marks to prevent your shell from interpreting the |.

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 geometry (e.g. -size 640x512 "image.rgb[320x256+50+50]"). Surround the image 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 document with +adjoin specified) may be written with the scene number included 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 specification in the file name and use the +adjoin option. For example,

    image%02d.miff

writes files image00.miff, image01.miff, etc. Only a single specification is allowed within an output filename. If more than one specification is present, it will be ignored. It is best to embed the scene number 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 convenient in the event you have too many image filenames to fit on the command line.

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 encoding. The latter can appear after the final group of input images.

This is a combined list of the command-line options used by the GraphicsMagick 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 <distance>{%}" means you can use the option "-fuzz 10" or "-fuzz 2%".
 

-adjoin

    image%02d.miff

-affine <matrix>

-antialias

-append

-asc-cdl <spec>

  redslope,redoffset,redpower:
  greenslope,greenoffset,greenpower:
  blueslope,blueoffset,bluepower:
  saturation

-authenticate <string>

-auto-orient

-average

-backdrop

-background <color>

-black-threshold red[,green][,blue][,opacity]

-blue-primary <x>,<y>

-blur <radius>{x<sigma>}

-border <width>x<height>

-bordercolor <color>

-borderwidth <geometry>

-box <color>

-channel <type>

-charcoal <factor>

-chop <width>x<height>{+-}<x>{+-}<y>{%}

-clip

    gm convert -clip -negate cockatoo.tif negated.tif

-coalesce

-colorize <value>

    gm convert input.ppm -modulate 115,0,100 \
              -colorize 7,21,50 output.ppm.

-colormap <type>

-colors <value>

-colorspace <value>

-comment <string>

     -comment "%m:%f %wx%h"

-compose <operator>

     Over
     In
     Out
     Atop
     Xor
     Plus
     Minus
     Add
     Subtract
     Difference
     Divide
     Multiply
     Bumpmap
     Copy
     CopyRed
     CopyGreen
     CopyBlue
     CopyOpacity
     CopyCyan
     CopyMagenta
     CopyYellow
     CopyBlack

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 values 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 operators can be used to perform reversible transformations.

Difference

The result of abs(change-image - base-image). This is useful 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 illuminated 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-image. Here the matte information is ignored.

CopyRed

The resulting image is the red channel in base-image replaced with the red channel in change-image. The other channels are copied untouched.

CopyGreen

The resulting image is the green channel in base-image replaced with the green channel in change-image. The other channels are copied untouched.

CopyBlue

The resulting image is the blue channel in base-image replaced 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 replaced with the cyan channel in change-image. The other channels are copied untouched. Use of this operator requires 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 replaced with the black channel in change-image. The other channels are copied untouched. Use of this operator requires 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>

-contrast

    gm convert rose: -contrast -contrast rose_c2.png

-convolve <kernel>

-create-directories

-crop <width>x<height>{+-}<x>{+-}<y>{%}

-cycle <amount>

-debug <events>

    gm convert -debug "Cache,Blob" rose: rose.png

-deconstruct

-define <key>{=<value>},...

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 endianness.

dpx:swap-samples={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 interchanged, or Cb and Cr interchanged due to an different interpretation 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.

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: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:optimize-coding={true|false}

Selects if huffman encoding should be used. Huffman encoding 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 sampling-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).

pdf:use-cropbox

If the pdf:use-cropbox flag is defined, then Ghostscript is requested to apply the PDF crop box.

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.

tiff:alpha={unspecified|associated|unassociated}

Specify the TIFF alpha channel type when reading or writing TIFF files, overriding the normal value. The default alpha channel type for new files is associated alpha. Existing 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 appear as if they were alpha composited on a black background when the matte channel is disabled. If the unassociated 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 exception to this is when Group3 or Group4 FAX compression is requested since FAX machines send data in bit-reversed order and therefore RFC 2301 recommends using reverse order.

tiff:group-three-options=<value>

If the tiff:group-three-options key is defined, GraphicsMagick will use it to set the group3 options tag when writing group3-compressed TIFF. Please see the TIFF specification for the usage of this tag. The default value is 4.

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, GraphicsMagick 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 obtained from the file's SMaxSampleValue tag (if present). The floating point data is currently not scanned in advance 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, GraphicsMagick 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 obtained 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 specification 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 image. 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 compression 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. Enables 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 internal 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 benefit 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.

    gm convert bilevel.tif -define ps:imagemask eps3:stencil.ps

-delay <1/100ths of a second>

-density <width>x<height>

-depth <value>

-descend

-despeckle

-displace <horizontal scale>x<vertical scale>

-display <host:display[.screen]>

-dispose <method>

    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>

-dither

-draw <string>

     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

     text            x0,y0 string

     gravity         NorthWest, North, NorthEast, West, Center,
                     East, SouthWest, South, or SouthEast

     rotate          degrees
     translate       dx,dy
     scale           sx,sy
     skewX           degrees
     skewY           degrees

     color           x0,y0 method
     matte           x0,y0 method

     -draw 'circle 100,100 150,150'

     -draw 'image Over 100,100 225,225 image.jpg'

     -draw 'text 100,100 "%m:%f %wx%h"'

     point
     replace
     floodfill
     filltoborder
     reset

-edge <radius>

-emboss <radius>

-encoding <type>

-endian <type>

-enhance

-equalize

-extent <width>x<height>{+-}<x>{+-}<y>

    gm convert infile.jpg -thumbnail 120x80 -background red -gravity center \
              -extent 140x100 outfile.jpg

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 obvious. 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 equivalent. 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 approximate 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, supports a matte channel, or is being enlarged, otherwise the Lanczos 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 transparent 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, TrueType, 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 extension is replaced with the image format type specified with -format. 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 option 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 special format characters: %b file size %c comment %d directory %e filename extension %f filename %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 minimum bit depth %r image type description %s scene number %t top of filename %u unique temporary filename %w width %x x resolution %y y resolution %# 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. The values of image type (%p) 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>" can be one of the following: * (print all Exif tags, in keyword=data format) ! (print all Exif tags, in tag_number data 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 Surround the format specification with quotation marks to prevent 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 command 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>{%}{@}{!}{^}{<}{>} preferred size and location of the Image window. By default, the window size is the image size and the location is chosen by you when it is mapped. 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 is resized while maintaining the aspect ratio of the image, but the resulting width or height are treated as minimum values rather than maximum values. Append an exclamation point to the geometry to force the image 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, the width assumes the value and the height is chosen to maintain the aspect ratio of the image. Similarly, if only the height is specified (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 specification. 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. Enclose the geometry specification in quotation marks to prevent the < or > from being interpreted by your shell as a file redirection. 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 image. 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 dimensions 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 image. 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 direction 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 colors) and of order 16 is 4096x4096 (16,777,216 colors). A special 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 default 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 attention 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 interlace (RGBRGBRGBRGBRGBRGB...), Line uses scanline interlacing (RRR...GGG...BBB...RRR...GGG...BBB...), and Plane uses plane interlacing (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 filename, 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. 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 maximum 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, producing a bi-level equivalent.

-level <black_point>{,<gamma>}{,<white_point>}{%}

-limit <type> <value>

-linewidth

-list <type>

-log <string>

     %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

    gm convert -debug coders -log "%u %m:%l %e" in.gif out.png

-loop <iterations>

-magnify

-magnify <factor>

-map <filename>

-map <type>

     best
     default
     gray
     red
     green
     blue

-mask <filename>

-matte

-mattecolor <color>

-maximum-error <limit>

-median <radius>

-metric <metric>

-minify <factor>

-mode <value>

-modulate brightness[,saturation[,hue]]

-monitor

-monochrome

-morph <frames>

-mosaic

    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}

-name

-negate

-noise <radius|type>

     Uniform
     Gaussian
     Multiplicative
     Impulse
     Laplacian
     Poisson
     Random (uniform distribution)

-noop

-normalize

-opaque <color>

-operator channel operator rvalue[%]

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 (approximately) stored in the specified number of bits without 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. Inverting 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 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 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 gaussian 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 laplacian noise according to the intensity specified by rvalue.

Noise-Multiplicative

Result is the current channel value modulated with multiplicative gaussian noise according to the intensity specified 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 according to the intensity specified by rvalue.

    gm convert in.bmp -operator red assign "50%" out.bmp

    gm convert in.bmp -operator gray threshold "50%" out.bmp

-ordered-dither <channeltype> <NxN>

-output-directory <directory>

-orient <orientation>

-page <width>x<height>{+-}<x>{+-}<y>{%}{!}{<}{>}

     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

-paint <radius>

-pause <seconds>

-pause <seconds>

-pen <color>

-ping

-pointsize <value>

-preview <type>

     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

-process <command>

-profile <filename>

    gm convert cockatoo.jpg exifdata.app1

+progress

-quality <value>

     0: none
     1: sub
     2: up
     3: average
     4: Paeth

-raise <width>x<height>

-random-threshold <channeltype> <LOWxHIGH>

-recolor <matrix>

  1 0 0
  0 1 0
  0 0 1

  -recolor "1 0 0, 0 1 0, 0 0 1"

  1 0 0 0
  0 1 0 0
  0 0 1 0
  0 0 0 1

  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

  0 0 1
  0 1 0
  1 0 0

  0.2126 0.7152 0.0722
  0.2126 0.7152 0.0722
  0.2126 0.7152 0.0722

  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>

-region <width>x<height>{+-}<x>{+-}<y>

-remote

-render

-repage <width>x<height>+xoff+yoff[!]

-resample <horizontal>x<vertical>

-resize <width>x<height>{%}{@}{!}{<}{>}

-roll {+-}<x>{+-}<y>

-rotate <degrees>{<}{>}

-sample <geometry>

-sampling-factor <horizontal_factor>x<vertical_factor>

-scale <geometry>

-scene <value>

-scenes <value-value>

    gm montage -scenes 5-7 image.miff montage.miff

    gm animate -scenes 0-12 image%02d.miff

-screen

-set <attribute> <value>

-segment <cluster threshold>x<smoothing threshold>

-shade <azimuth>x<elevation>

-shadow <radius>{x<sigma>}

-shared-memory

-sharpen <radius>{x<sigma>}

-shave <width>x<height>{%}

-shear <x degrees>x<y degrees>

-silent

-size <width>x<height>{+offset}

     192x128
     384x256
     768x512
     1536x1024
     3072x2048

-snaps <value>

-solarize <factor>