From 801e2ff28a4292868da219979ae197c14d46d6d4 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Bastien=20ROUCARI=C3=88S?= Date: Wed, 14 Dec 2016 15:26:15 +0100 Subject: Fix www/command-line-options.html Forwarded: Yes, https://github.com/ImageMagick/ImageMagick/issues/304 --- www/command-line-options.html | 721 +++++++++++++++++++++++++++++++++- 1 file changed, 720 insertions(+), 1 deletion(-) diff --git a/www/command-line-options.html b/www/command-line-options.html index 3280c00f3c..f347420207 100644 --- a/www/command-line-options.html +++ b/www/command-line-options.html @@ -1856,7 +1856,725 @@ is case-dependent.

Such settings are global in scope, and affect all images and operations.

-

See ImageMagick Defines for a list of recognized defines. For example:

+

The following definitions are just some of the artifacts that are +available:

+ + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + + +
auto-threshold:verbosereturn derived threshold as the auto-threshold:threshold image property.
bmp3:alpha=true|falseinclude any alpha channel when writing in the BMP image format.
bmp:format=value valid values are bmp2, bmp3, + and bmp4. This option can be useful when the + method of prepending "BMP2:" to the output filename is inconvenient or + is not available, such as when using the mogrify utility.
bmp:subtype=RGB555|RGB565|ARGB4444|ARGB1555BMP channel depth subtypes. Only support in BMP (BMP4). BMP3 and BMP2 do not + contain header fields to support these options.
colorspace:auto-grayscale=on|offprevent automatic conversion to grayscale inside coders that support + grayscale. This should be accompanied by -type truecolor. PNG and TIF do + not need this define. With PNG, just use PNG24:image. With TIF, just use + -type truecolor. JPG and PSD will need this define.
complex:snr=valueSet the divide SNR constant-complex
compose:args=argumentsSets certain compose argument values when using convert ... -compose ... + -composite. See Image Composition
compose:clamp=on|offset each pixel whose value is below zero to zero and any the pixel whose value is above the quantum range to the quantum range (e.g. 65535) otherwise the pixel value remains unchanged. Define supported in ImageMagick 6.9.1-3 and above.
compose:outside-overlay=true|falseSome -compose methods can modify the 'destination' image outside the overlay area. It is disabled by default.
connected-components:area-threshold=valueMerges any region with area smaller than value into its surrounding region or largest neighbor.
connected-components:keep=list-of-idsComma and/or hyphenated list of id values to keep in the output. Supported in Imagemagick 6.9.3-0.
connected-components:mean-color=trueChanges the output image from id values to mean color values. Supported in Imagemagick 6.9.2-8.
connected-components:remove=list-of-idsComma and/or hyphenated list of id values to remove from the output. Supported in Imagemagick 6.9.2-9.
connected-components:verbose=trueLists id, bounding box, centroid, area, mean color for each region.
convolve:scale={kernel_scale}[!^] [,{origin_addition}] [%]Defines the kernel scaling. The special flag ! automatically scales to + full dynamic range. The ! flag can be used in combination with a factor or + percent. The factor or percent is then applied after the automatic scaling. + An example is 50%!. This produces a result 50% darker than full dynamic + range scaling. The ^ flag assures the kernel is 'zero-summing', for + example when some values are positive and some are negative as in edge + detection kernels. The origin addition adds that value to the center + pixel of the kernel. This produces an effect that is like adding the image + that many times to the result of the filtered image. The typical value + is 1 so that the original image is added to the result of the convolution. + The default is 0.
convolve:showKernel=1Outputs (to 'standard error') all the information about a specified -morphology convolve kernel.
dcm:display-range=resetSets the display range to the minimum and maximum pixel values for the + DCM image format.
dds:cluster-fit=true|falseEnables the dds cluster-fit.
dds:compression=dxt1|dxt5|noneSets the dds compression.
dds:mipmaps=valueSets the dds number of mipmaps.
dds:weight-by-alpha=true|falseEnables the dds alpha weighting.
deskew:auto-crop=trueauto crop the image after deskewing.
delegate:bimodal=trueSpecifies direct conversion from Postscript to PDF.
distort:scale=valueSets the output scaling factor for use with -distort
distort:viewport=WxH+X+YSets the viewport for use with -distort
dither:diffusion-amount=X%Sets the amount of diffusion to use with Floyd-Steinberg diffusion
dot:layout-engine=valueSpecifies the layout engine for the DOT image format (e.g. + neato).
exr:color-type=valueSpecifies the color type for the EXR format: RGB, RGBA, YC, YCA, Y, YA, R, G, B, A).
filter:option=valueSet a filter option for use with -resize. + See -filter for details.
fourier:normalize=inverseSets the location for the FFT/IFT normalization as use by + +-fft and +-ift. The default is + forward.
h:format=valueSet the image encoding format use when writing a C-style header. + format can be any output format supported by ImageMagick + except for h and magick. If this + option is omitted, the default is GIF for PseudoClass + images and PNM for DirectClass images. +
hough-lines:accumulator=trueReturns the accumulator image in addition to the lines image
icon:auto-resizeAutomatically stores multiple sizes when writing an ico image + (requires a 256x256 input image).
identify:locate=minimum|maximumLocates the coordinates of one or more image minimum or maximum
identify:limit=numberLocates the coordinates for the number of minima or maxima specified
jp2:layer-number=valueSets the maximum number of quality layers to decode. Same for JPT, JC2, + and J2K
jp2:number-resolutions=valueSets the number of resolutions to encode.Same for JPT, JC2, and J2K
jp2:progression-order=valuechoose from LRCP, RLCP, RPCL, PCRL or CPRL. Same for JPT, JC2, and + J2K
jp2:quality=value,value...Sets the quality layer PSNR, given in dB. The order is from left to + right in ascending order. The default is a single lossless quality layer. + Same for JPT, JC2, and J2K
jp2:rate=valueSpecify 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. A quality setting of 75 + results in a rate value of 0.06641. Same for JPT, JC2, and J2K
jp2:reduce-factor=valueSets the number of highest resolution levels to be discarded.Same for + JPT, JC2, and J2K
jpeg:block-smoothing=on|off
jpeg:colors=valueSet the desired number of colors and let the JPEG encoder do the + quantizing.
jpeg:dct-method=valueChoose from default, fastest, + float, ifast, and islow.
jpeg:extent=valueRestrict the maximum JPEG file size, for example -define + jpeg:extent=400KB. The JPEG encoder will search for the highest + compression quality level that results in an output file that does not + exceed the value. The -quality option also will be respected + starting with version 6.9.2-5. Between 6.9.1-0 and 6.9.2-4, add -quality + 100 in order for the jpeg:extent to work properly. Prior to 6.9.1-0, the + -quality setting was ignored.
jpeg:fancy-upsampling=on|off
jpeg:optimize-coding=on|off
jpeg:q-table=table
jpeg:sampling-factor=sampling-factor-string
jpeg:size=geometrySet the size hint of a JPEG image, for + example, -define jpeg:size=128x128. + It is most useful for increasing performance and reducing the memory + requirements when reducing the size of a large JPEG image.
json:featuresincludes features in verbose information
json:limit
json:locate
json:momentsincludes image moments in verbose information
magick:format=valueSet the image encoding format use when writing a C-style header. + This is the same as "h:format=format" described above.
mng:need-cacheoffturn playback caching off for streaming MNG.
morphology:compose=compose-methodSpecifies how to merge results generated by multiple-morphology kernel. The default is none. One + typical value is 'lighten' as used, for example, with the sobel edge + kernels.
morphology:showKernel=1Outputs (to 'standard error') all the information about a generated -morphology kernel.
pcl:fit-to-page=true
pdf:fit-page=geometry geometry specifies the scaling dimensions for resizing when the PDF is + being read. The geometry is either WxH{%} or page size. No offsets are + allowed. (introduced in IM 6.8.8-8)
pdf:fit-to-page=true
pdf:use-cropbox=true
pdf:use-trimbox=true
pdf:stop-on-error=true
pdf:pdf:page-direction=right-to-left
png:bit-depth=value
png:color-type=valuedesired bit-depth and color-type for PNG output. You can force the PNG encoder to use a different bit-depth and color-type than it would have + normally selected, but only if this does not cause any loss of image + quality. Any attempt to reduce image quality is treated as an error and no + PNG file is written. E.g., if you have a 1-bit black-and-white image, you + can use these "defines" to cause it to be written as an 8-bit grayscale, + indexed, or even a 64-bit RGBA. But if you have a 16-million color image, + you cannot force it to be written as a grayscale or indexed PNG. If you + wish to do this, you must use the appropriate -depth, + -colors, or -type directives to + reduce the image quality prior to using the PNG encoder. Note that in + indexed PNG files, "bit-depth" refers to the number of bits per index, + which can be 1, 2, 4, or 8. In such files, the color samples always have + 8-bit depth.
png:compression-filter=value valid values are 0 through 9. 0-4 are the corresponding PNG filters, + 5 means adaptive filtering except for images with a colormap, 6 means + adaptive filtering for all images, 7 means MNG "loco" compression, 8 means + Z_RLE strategy with adaptive filtering, and 9 means Z_RLE strategy with no + filtering.
png:compression-level=value valid values are 0 through 9, with 0 providing the least but fastest + compression and 9 usually providing the best and always the slowest.
png:compression-strategy=value valid values are 0 through 4, meaning default, filtered, huffman_only, + rle, and fixed ZLIB compression strategy. If you are using an old zlib + that does not support Z_RLE (before 1.2.0) or Z_FIXED (before 1.2.2.2), + values 3 and 4, respectively, will use the zlib default strategy + instead.
png:format=value valid values are png8, png24, + png32, png48, + png64, and png00. + This property can be useful for specifying + the specific PNG format to be used, when the usual method of prepending the + format name to the output filename is inconvenient, such as when writing + a PNG-encoded ICO file or when using mogrify. + Value = png8 reduces the number of colors to 256, + only one of which may be fully transparent, if necessary. The other + values do not force any reduction of quality; it is an error to request + a format that cannot represent the image data without loss (except that + it is allowed to reduce the bit-depth from 16 to 8 for all formats). + Value = png24 and png48 + allow transparency, only if a single color is fully transparent and that + color does not also appear in an opaque pixel; such transparency is + written in a PNG tRNS chunk. + Value = png00 causes the image to inherit its + color-type and bit-depth from the input image, if the input was also + a PNG.
png:exclude-chunk=value
png:include-chunk=valueancillary chunks to be excluded from or included in PNG output. + +

The value can be the name of a PNG chunk-type such + as bKGD, a comma-separated list of chunk-names + (which can include the word date, the word + all, or the word none). + Although PNG chunk-names are case-dependent, you can use all lowercase + names if you prefer.

+ +

The "include-chunk" and "exclude-chunk" lists only affect the behavior + of the PNG encoder and have no effect on the PNG decoder.

+ +

As a special case, if the sRGB chunk is excluded and + the gAMA chunk is included, the gAMA chunk will + only be written if gamma is not 1/2.2, since most decoders assume + sRGB and gamma=1/2.2 when no colorspace information is included in + the PNG file. Because the list is processed from left to right, you + can achieve this with a single define:

+ +
-define png:include-chunk=none,gAMA
+
+ +

As a special case, if the sRGB chunk is not excluded and + the PNG encoder recognizes that the image contains the sRGB ICC profile, + the PNG encoder will write the sRGB chunk instead of the + entire ICC profile. To force the PNG encoder to write the sRGB + profile as an iCCP chunk in the output PNG instead of the + sRGB chunk, exclude the sRGB chunk.

+ +

The critical PNG chunks IHDR, PLTE, + IDAT, and IEND cannot be excluded. Any such + entries appearing in the list will be ignored.

+ +

If the ancillary PNG tRNS chunk is excluded and the + image has transparency, the PNG colortype is forced to be 4 or 6 + (GRAY_ALPHA or RGBA). If the image is not transparent, then the + tRNS chunk isn't written anyhow, and there is no effect + on the PNG colortype of the output image.

+ +

The -strip option does the equivalent of the + following for PNG output:

+ +
-define png:exclude-chunk=EXIF,iCCP,iTXt,sRGB,tEXt,zCCP,zTXt,date
+
+ +

The default behavior is to include all known PNG ancillary chunks + plus ImageMagick's private vpAg ("virtual page") chunk, + and to exclude all PNG chunks that are unknown to ImageMagick, + regardless of their PNG "copy-safe" status as described in the + PNG specification.

+ +

Any chunk names that are not known to ImageMagick are ignored + if they appear in either the "include-chunk" or "exclude-chunk" list. + The ancillary chunks currently known to ImageMagick are + bKGD, cHRM, gAMA, iCCP, + oFFs, pHYs, sRGB, tEXt, + tRNS, vpAg, and zTXt.

+ +

You can also put date in the list to include or exclude + the "Date:create" and "Date:modify" text chunks that ImageMagick normally + inserts in the output PNG.

png:ignore-crc[=true]When you know your image has no CRC or ADLER32 errors, this can speed up + decoding. It is also helpful in debugging bug reports from "fuzzers"
png:preserve-colormap[=true]Use the existing image->colormap. Normally the PNG encoder will + try to optimize the palette, eliminating unused entries and putting + the transparent colors first. If this flag is set, that behavior + is suppressed.
png:preserve-iCCP[=true]By default, the PNG decoder and encoder examine any ICC profile + that is present, either from an iCCP chunk in the PNG + input or supplied via an option, and if the profile is recognized + to be the sRGB profile, converts it to the sRGB chunk. + You can use -define png:preserve-iCCP to prevent + this from happening; in such cases the iCCP chunk + will be read or written and no sRGB chunk will be + written. There are some ICC profiles that claim to be sRGB but + have various errors that cause them to be rejected by libpng16; such + profiles are recognized anyhow and converted to the sRGB + chunk, but are rejected if the -define png:preserve-iCCP + is present. Note that not all "sRGB" ICC profiles are recognized + yet; we will add them to the list as we encounter them.
png:swap-bytes[=true]The PNG specification requires that any multi-byte integers be stored in network byte order (MSB-LSB endian). This option allows you to + fix any invalid PNG files that have 16-bit samples stored incorrectly + in little-endian order (LSB-MSB). The "-define png:swap-bytes" option + must appear before the input filename on the commandline. The swapping + is done during the libpng decoding operation.
profile:skip=name1,name2,...Skip the named profile[s] when reading the image. Use skip="*" to + skip all named profiles in the image. Many named profiles exist, + including ICC, EXIF, APP1, IPTC, XMP, and others.
ps:imagemaskIf 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.
psd:additional-info=all|selectiveThis option should only be used when converting from a PSD file to another PSD file. This should be placed after the image is read. The two options are 'all' and 'selective'. The 'selective' option will preserve all additional information that is not related to the geometry of the image. The 'all' option should only be used when the geometry of the image has not been changed. This option is helpful when transferring non-simple layers, such as adjustment layers from the input PSD file to the output PSD file. If this option is not used, the additional information will not be preserved. This define is available as of Imagemagick version 6.9.5-8. +
psd:alpha-unblend=offDisables new automatic un-blending of transparency with the base image + for the flattened layer 0 before adding the alpha channel to the output + image. This define must be placed before the input psd image. (Available + as of IM 6.9.2.5). The automatic un-blending is new to IM 6.9.2.5 and + prevents the transparency from being applied twice in the output + image. This option should be set before reading the image.
psd:preserve-opacity-mask=trueThis option should only be used when converting from a PSD file to another PSD file. It will preserve the opacity mask of a layer and add it back to the layer when the image is saved. Setting this to 'true' will enable this feature. This define is available as of Imagemagick version 6.9.5-10. +
preserve-timestamp={True, False}Preserve file timestamp (mogrify only).
q-table=quantization-table.xmlCustom JPEG quantization tables.
quantum:format=typeSet the type to floating-point to specify a floating-point + format for raw files (e.g. GRAY:) or for MIFF and TIFF images in HDRI mode + to preserve negative values. If -depth 16 is + included, the result is a single precision floating point format. + If -depth 32 is included, the result is + double precision floating point format. For signed pixel data, use -define quantum:format=signed.
quantum:polarity=photometric-interpretationSet the photometric-interpretation of an image (typically for TIFF + image file format) to either min-is-black (default) or + min-is-white.
sample:offset=geometryLocation of the sampling point within the sub-region being sampled, + expressed as percentages (see -sample).
showKernel=1Outputs (to 'standard error') all the information about a generated -morphology kernel.
stream:buffer-size=valueSet the stream buffer size. Select 0 for unbuffered I/O.
tiff:alpha=associated|unassociated|unspecifiedSpecify the alpha extra samples as associated, unassociated or + unspecified
tiff:endian=msb|lsb
tiff:exif-properties=trueEnables reading the EXIF properties.
tiff:fill-order=msb|lsb
tiff:ignore-layers=trueIgnores the photoshop layers.
tiff:ignore-tags=comma-separate-list-of-tag-IDsAllows one or more tag ID values to be ignored.
tiff:predictor=[1, 2 or 3]A mathematical operator that is applied to the image data before an encoding scheme is applied. The general idea is that subsequent pixels of an image resemble each other. Thus, substracting the information from a pixel that is already contained in previous one is likely to reduce its information density considerably and aid subsequent compression. 1 = No prediction scheme used before coding. 2 = Horizontal differencing. 3 = Floating point horizontal differencing.
tiff:rows-per-strip=valueSets the number of rows per strip
tiff:tile-geometry=WxHSets the tile size for pyramid tiffs. Requires the suffix + PTIF: before the outputname
type:features=string
type:hinting=off
txt:compliance=cssThe first argument of a hsl() color is integer, not percentage
x:screen=trueObtains the image from the root window
x:silent=trueTurns off the beep when importing an image
+ +

For example, to create a postscript file that will render only the black +pixels of a bilevel image, use:

convert bilevel.tif -define ps:imagemask eps3:stencil.ps
@@ -7457,3 +8175,4 @@ but with strict boolean masking.