MAN page from CentOS 7 mozjpeg-master-20170714.2.x86_64.rpm


Section: User Commands (1)
Updated: 18 March 2017


cjpeg - compress an image file to a JPEG file 





cjpegcompresses the named image file, or the standard input if no file isnamed, and produces a JPEG/JFIF file on the standard output.The currently supported input file formats are: PPM (PBMPLUS colorformat), PGM (PBMPLUS grayscale format), BMP, Targa, and RLE (Utah RasterToolkit format). (RLE is supported only if the URT library is available.) 


All switch names may be abbreviated; for example,-grayscalemay be written-grayor-gr.Most of the "basic" switches can be abbreviated to as little as one letter.Upper and lower case are equivalent (thus-BMPis the same as-bmp).British spellings are also accepted (e.g.,-greyscale),though for brevity these are not mentioned below.

The basic switches are:

-quality N[,...]
Scale quantization tables to adjust image quality. Quality is 0 (worst) to100 (best); default is 75. (See below for more info.)
Create monochrome JPEG file from color input. Be sure to use this switch whencompressing a grayscale BMP file, becausecjpegisn't bright enough to notice whether a BMP file uses only shades of gray.By saying-grayscale,you'll get a smaller JPEG file that takes less time to process.
Create RGB JPEG file.Using this switch suppresses the conversion from RGBcolorspace input to the default YCbCr JPEG colorspace.
Perform optimization of entropy encoding parameters. Without this, defaultencoding parameters are used.-optimizeusually makes the JPEG file a little smaller, butcjpegruns somewhat slower and needs much more memory. Image quality and speed ofdecompression are unaffected by-optimize.
Create progressive JPEG file (see below).
Input file is Targa format. Targa files that contain an "identification"field will not be automatically recognized bycjpeg;for such files you must specify-targato makecjpegtreat the input as Targa format.For most Targa files, you won't need this switch.

The-qualityswitch lets you trade off compressed file size against quality of thereconstructed image: the higher the quality setting, the larger the JPEG file,and the closer the output image will be to the original input. Normally youwant to use the lowest quality setting (smallest file) that decompresses intosomething visually indistinguishable from the original image. For thispurpose the quality setting should generally be between 50 and 95 (the defaultis 75) for photographic images. If you see defects at-quality75, then go up 5 or 10 counts at a time until you are happy with the outputimage. (The optimal setting will vary from one image to another.)

-quality100 will generate a quantization table of all 1's, minimizing loss in thequantization step (but there is still information loss in subsampling, as wellas roundoff error.) For most images, specifying a quality value aboveabout 95 will increase the size of the compressed file dramatically, and whilethe quality gain from these higher quality values is measurable (using metricssuch as PSNR or SSIM), it is rarely perceivable by human vision.

In the other direction, quality values below 50 will produce very small filesof low image quality. Settings around 5 to 10 might be useful in preparing anindex of a large image library, for example. Try-quality2 (or so) for some amusing Cubist effects. (Note: qualityvalues below about 25 generate 2-byte quantization tables, which areconsidered optional in the JPEG standard.cjpegemits a warning message when you give such a quality value, because someother JPEG programs may be unable to decode the resulting file. Use-baselineif you need to ensure compatibility at low quality values.)

The -quality option has been extended in this version of cjpeg tosupport separate quality settings for luminance and chrominance (or, ingeneral, separate settings for every quantization table slot.) The principleis the same as chrominance subsampling: since the human eye is more sensitiveto spatial changes in brightness than spatial changes in color, the chrominancecomponents can be quantized more than the luminance components withoutincurring any visible image quality loss. However, unlike subsampling, thisfeature reduces data in the frequency domain instead of the spatial domain,which allows for more fine-grained control. This option is useful inquality-sensitive applications, for which the artifacts generated bysubsampling may be unacceptable.

The -quality option accepts a comma-separated list of parameters, whichrespectively refer to the quality levels that should be assigned to thequantization table slots. If there are more q-table slots than parameters,then the last parameter is replicated. Thus, if only one quality parameter isgiven, this is used for both luminance and chrominance (slots 0 and 1,respectively), preserving the legacy behavior of cjpeg v6b and prior.More (or customized) quantization tables can be set with the -qtablesoption and assigned to components with the -qslots option (see the"wizard" switches below.)

JPEG files generated with separate luminance and chrominance quality are fullycompliant with standard JPEG decoders.

CAUTION:For this setting to be useful, be sure to pass an argument of -sample 1x1to cjpeg to disable chrominance subsampling. Otherwise, the defaultsubsampling level (2x2, AKA "4:2:0") will be used.

The-progressiveswitch creates a "progressive JPEG" file. In this type of JPEG file, the datais stored in multiple scans of increasing quality. If the file is beingtransmitted over a slow communications link, the decoder can use the firstscan to display a low-quality image very quickly, and can then improve thedisplay with each subsequent scan. The final image is exactly equivalent to astandard JPEG file of the same quality setting, and the total file size isabout the same --- often a little smaller.

Switches for advanced users:

Use arithmetic coding.Caution:arithmetic coded JPEG is not yet widely implemented, so many decoders will beunable to view an arithmetic coded JPEG file at all.
-dct int
Use integer DCT method (default).
-dct fast
Use fast integer DCT (less accurate).In libjpeg-turbo, the fast method is generally about 5-15% faster than the intmethod when using the x86/x86-64 SIMD extensions (results may vary with otherSIMD implementations, or when using libjpeg-turbo without SIMD extensions.)For quality levels of 90 and below, there should be little or no perceptibledifference between the two algorithms. For quality levels above 90, however,the difference between the fast and the int methods becomes more pronounced.With quality=97, for instance, the fast method incurs generally about a 1-3 dBloss (in PSNR) relative to the int method, but this can be larger for someimages. Do not use the fast method with quality levels above 97. Thealgorithm often degenerates at quality=98 and above and can actually produce amore lossy image than if lower quality levels had been used. Also, inlibjpeg-turbo, the fast method is not fully accelerated for quality levelsabove 97, so it will be slower than the int method.
-dct float
Use floating-point DCT method.The float method is mainly a legacy feature. It does not produce significantlymore accurate results than the int method, and it is much slower. The floatmethod may also give different results on different machines due to varyingroundoff behavior, whereas the integer methods should give the same results onall machines.
-restart N
Emit a JPEG restart marker every N MCU rows, or every N MCU blocks if "B" isattached to the number.-restart 0(the default) means no restart markers.
-smooth N
Smooth the input image to eliminate dithering noise. N, ranging from 1 to100, indicates the strength of smoothing. 0 (the default) means no smoothing.
-maxmemory N
Set limit for amount of memory to use in processing large images. Value isin thousands of bytes, or millions of bytes if "M" is attached to thenumber. For example,-max 4mselects 4000000 bytes. If more space is needed, an error will occur.
-outfile name
Send output image to the named file, not to standard output.
Compress to memory instead of a file. This feature was implemented mainly as away of testing the in-memory destination manager (jpeg_mem_dest()), but it isalso useful for benchmarking, since it reduces the I/O overhead.
Enable debug printout. More-v'sgive more output. Also, version information is printed at startup.
Same as-verbose.
Print version information and exit.

The-restartoption inserts extra markers that allow a JPEG decoder to resynchronize aftera transmission error. Without restart markers, any damage to a compressedfile will usually ruin the image from the point of the error to the end of theimage; with restart markers, the damage is usually confined to the portion ofthe image up to the next restart marker. Of course, the restart markersoccupy extra space. We recommend-restart 1for images that will be transmitted across unreliable networks such as Usenet.

The-smoothoption filters the input to eliminate fine-scale noise. This is often usefulwhen converting dithered images to JPEG: a moderate smoothing factor of 10 to50 gets rid of dithering patterns in the input file, resulting in a smallerJPEG file and a better-looking image. Too large a smoothing factor willvisibly blur the image, however.

Switches for wizards:

Force baseline-compatible quantization tables to be generated. This clampsquantization values to 8 bits even at low quality settings. (This switch ispoorly named, since it does not ensure that the output is actually baselineJPEG. For example, you can use-baselineand-progressivetogether.)
-qtables file
Use the quantization tables given in the specified text file.
-qslots N[,...]
Select which quantization table to use for each color component.
-sample HxV[,...]
Set JPEG sampling factors for each color component.
-scans file
Use the scan script given in the specified text file.

The "wizard" switches are intended for experimentation with JPEG. If youdon't know what you are doing, don't use them. These switches aredocumented further in the file wizard.txt. 


This example compresses the PPM file foo.ppm with a quality factor of60 and saves the output as foo.jpg:

cjpeg -quality60 foo.ppm>foo.jpg


Color GIF files are not the ideal input for JPEG; JPEG is really intended forcompressing full-color (24-bit) images. In particular, don't try to convertcartoons, line drawings, and other images that have only a few distinctcolors. GIF works great on these, JPEG does not. If you want to convert aGIF to JPEG, you should experiment withcjpeg's-qualityand-smoothoptions to get a satisfactory conversion.-smooth 10or so is often helpful.

Avoid running an image through a series of JPEG compression/decompressioncycles. Image quality loss will accumulate; after ten or so cycles the imagemay be noticeably worse than it was after one cycle. It's best to use alossless format while manipulating an image, then convert to JPEG format whenyou are ready to file the image away.

The-optimizeoption tocjpegis worth using when you are making a "final" version for posting or archiving.It's also a win when you are using low quality settings to make very smallJPEG files; the percentage improvement is often a lot more than it is onlarger files. (At present,-optimizemode is always selected when generating progressive JPEG files.) 


If this environment variable is set, its value is the default memory limit.The value is specified as described for the-maxmemoryswitch.JPEGMEMoverrides the default value specified when the program was compiled, anditself is overridden by an explicit-maxmemory.


Wallace, Gregory K. "The JPEG Still Picture Compression Standard",Communications of the ACM, April 1991 (vol. 34, no. 4), pp. 30-44. 


Independent JPEG Group

This file was modified by The libjpeg-turbo Project to include only informationrelevant to libjpeg-turbo, to wordsmith certain sections, and to describefeatures not present in libjpeg. 


Support for GIF input files was removed in cjpeg v6b due to concerns overthe Unisys LZW patent. Although this patent expired in 2006, cjpeg stilllacks GIF support, for these historical reasons. (Conversion of GIF files toJPEG is usually a bad idea anyway, since GIF is a 256-color format.)

Not all variants of BMP and Targa file formats are supported.

The-targaswitch is not a bug, it's a feature. (It would be a bug if the Targa formatdesigners had not been clueless.)




This document was created byman2html,using the manual pages.