Mercurial > hgrepos > Python2 > PyMuPDF

comparison mupdf-source/thirdparty/libjpeg/usage.txt @ 2:b50eed0cc0ef upstream

Find changesets by keywords (author, files, the commit message), revision number or hash, or revset expression.
ADD: MuPDF v1.26.7: the MuPDF source as downloaded by a default build of PyMuPDF 1.26.4. The directory name has changed: no version number in the expanded directory now.
author Franz Glasner <fzglas.hg@dom66.de>
date Mon, 15 Sep 2025 11:43:07 +0200
parents
children
comparison
equal deleted inserted replaced
1:1d09e1dec1d9 2:b50eed0cc0ef
1 USAGE instructions for the Independent JPEG Group's JPEG software
2 =================================================================
3
4 This file describes usage of the JPEG conversion programs cjpeg and djpeg,
5 as well as the utility programs jpegtran, rdjpgcom and wrjpgcom. (See
6 the other documentation files if you wish to use the JPEG library within
7 your own programs.)
8
9 If you are on a Unix machine you may prefer to read the Unix-style manual
10 pages in files cjpeg.1, djpeg.1, jpegtran.1, rdjpgcom.1, wrjpgcom.1.
11
12
13 INTRODUCTION
14
15 These programs implement JPEG image encoding, decoding, and transcoding.
16 JPEG (pronounced "jay-peg") is a standardized compression method for
17 full-color and grayscale images.
18
19
20 GENERAL USAGE
21
22 We provide two programs, cjpeg to compress an image file into JPEG format,
23 and djpeg to decompress a JPEG file back into a conventional image format.
24
25 On Unix-like systems, you say:
26 cjpeg [switches] [imagefile] >jpegfile
27 or
28 djpeg [switches] [jpegfile] >imagefile
29 The programs read the specified input file, or standard input if none is
30 named. They always write to standard output (with trace/error messages to
31 standard error). These conventions are handy for piping images between
32 programs.
33
34 On most non-Unix systems, you say:
35 cjpeg [switches] imagefile jpegfile
36 or
37 djpeg [switches] jpegfile imagefile
38 i.e., both the input and output files are named on the command line. This
39 style is a little more foolproof, and it loses no functionality if you don't
40 have pipes. (You can get this style on Unix too, if you prefer, by defining
41 TWO_FILE_COMMANDLINE when you compile the programs; see install.txt.)
42
43 You can also say:
44 cjpeg [switches] -outfile jpegfile imagefile
45 or
46 djpeg [switches] -outfile imagefile jpegfile
47 This syntax works on all systems, so it is useful for scripts.
48
49 The currently supported image file formats are: PPM (PBMPLUS color format),
50 PGM (PBMPLUS grayscale format), BMP, GIF, Targa, and RLE (Utah Raster Toolkit
51 format). (RLE is supported only if the URT library is available, which it
52 isn't on most non-Unix systems.) cjpeg recognizes the input image format
53 automatically, with the exception of some Targa-format files. You have to
54 tell djpeg which format to generate.
55
56 JPEG files are in the standard JFIF file format. There are other,
57 less widely used JPEG-based file formats, but we don't support them.
58
59 All switch names may be abbreviated; for example, -grayscale may be written
60 -gray or -gr. Most of the "basic" switches can be abbreviated to as little as
61 one letter. Upper and lower case are equivalent (-BMP is the same as -bmp).
62 British spellings are also accepted (e.g., -greyscale), though for brevity
63 these are not mentioned below.
64
65
66 CJPEG DETAILS
67
68 The basic command line switches for cjpeg are:
69
70 -quality N[,...] Scale quantization tables to adjust image quality.
71 Quality is 0 (worst) to 100 (best); default is 75.
72 (See below for more info.)
73
74 -grayscale Create monochrome JPEG file from color input.
75 Be sure to use this switch when compressing a grayscale
76 BMP or GIF file, because cjpeg isn't bright enough to
77 notice whether a BMP or GIF file uses only shades of
78 gray. By saying -grayscale, you'll get a smaller
79 JPEG file that takes less time to process.
80
81 -rgb Create RGB JPEG file.
82 Using this switch suppresses the conversion from RGB
83 colorspace input to the default YCbCr JPEG colorspace.
84 You can use this switch in combination with the
85 -block N switch (see below) for lossless JPEG coding.
86 See also the -rgb1 switch below.
87
88 -optimize Perform optimization of entropy encoding parameters.
89 Without this, default encoding parameters are used.
90 -optimize usually makes the JPEG file a little smaller,
91 but cjpeg runs somewhat slower and needs much more
92 memory. Image quality and speed of decompression are
93 unaffected by -optimize.
94
95 -progressive Create progressive JPEG file (see below).
96
97 -scale M/N Scale the output image by a factor M/N. Currently
98 supported scale factors are M/N with all N from 1 to
99 16, where M is the destination DCT size, which is 8 by
100 default (see -block N switch below).
101
102 -targa Input file is Targa format. Targa files that contain
103 an "identification" field will not be automatically
104 recognized by cjpeg; for such files you must specify
105 -targa to make cjpeg treat the input as Targa format.
106 For most Targa files, you won't need this switch.
107
108 The -quality switch lets you trade off compressed file size against quality of
109 the reconstructed image: the higher the quality setting, the larger the JPEG
110 file, and the closer the output image will be to the original input. Normally
111 you want to use the lowest quality setting (smallest file) that decompresses
112 into something visually indistinguishable from the original image. For this
113 purpose the quality setting should be between 50 and 95; the default of 75 is
114 often about right. If you see defects at -quality 75, then go up 5 or 10
115 counts at a time until you are happy with the output image. (The optimal
116 setting will vary from one image to another.)
117
118 -quality 100 will generate a quantization table of all 1's, minimizing loss
119 in the quantization step (but there is still information loss in subsampling,
120 as well as roundoff error). This setting is mainly of interest for
121 experimental purposes. Quality values above about 95 are NOT recommended for
122 normal use; the compressed file size goes up dramatically for hardly any gain
123 in output image quality.
124
125 In the other direction, quality values below 50 will produce very small files
126 of low image quality. Settings around 5 to 10 might be useful in preparing an
127 index of a large image library, for example. Try -quality 2 (or so) for some
128 amusing Cubist effects. (Note: quality values below about 25 generate 2-byte
129 quantization tables, which are considered optional in the JPEG standard.
130 cjpeg emits a warning message when you give such a quality value, because some
131 other JPEG programs may be unable to decode the resulting file. Use -baseline
132 if you need to ensure compatibility at low quality values.)
133
134 The -quality option has been extended in IJG version 7 for support of separate
135 quality settings for luminance and chrominance (or in general, for every
136 provided quantization table slot). This feature is useful for high-quality
137 applications which cannot accept the damage of color data by coarse
138 subsampling settings. You can now easily reduce the color data amount more
139 smoothly with finer control without separate subsampling. The resulting file
140 is fully compliant with standard JPEG decoders.
141 Note that the -quality ratings refer to the quantization table slots, and that
142 the last value is replicated if there are more q-table slots than parameters.
143 The default q-table slots are 0 for luminance and 1 for chrominance with
144 default tables as given in the JPEG standard. This is compatible with the old
145 behaviour in case that only one parameter is given, which is then used for
146 both luminance and chrominance (slots 0 and 1). More or custom quantization
147 tables can be set with -qtables and assigned to components with -qslots
148 parameter (see the "wizard" switches below).
149 CAUTION: You must explicitly add -sample 1x1 for efficient separate color
150 quality selection, since the default value used by library is 2x2!
151
152 The -progressive switch creates a "progressive JPEG" file. In this type of
153 JPEG file, the data is stored in multiple scans of increasing quality. If the
154 file is being transmitted over a slow communications link, the decoder can use
155 the first scan to display a low-quality image very quickly, and can then
156 improve the display with each subsequent scan. The final image is exactly
157 equivalent to a standard JPEG file of the same quality setting, and the total
158 file size is about the same --- often a little smaller.
159
160 Switches for advanced users:
161
162 -arithmetic Use arithmetic coding.
163 CAUTION: arithmetic coded JPEG is not yet widely
164 implemented, so many decoders will be unable to
165 view an arithmetic coded JPEG file at all.
166
167 -block N Set DCT block size. All N from 1 to 16 are possible.
168 Default is 8 (baseline format).
169 Larger values produce higher compression,
170 smaller values produce higher quality
171 (exact DCT stage possible with 1 or 2; with the
172 default quality of 75 and default quantization tables
173 the DCT+Quantization stage is lossless for N=1).
174 CAUTION: An implementation of the JPEG SmartScale
175 extension is required for this feature. SmartScale
176 enabled JPEG is not yet widely implemented, so many
177 decoders will be unable to view a SmartScale extended
178 JPEG file at all.
179
180 -rgb1 Create RGB JPEG file with reversible color transform.
181 Works like the -rgb switch (see above) and inserts a
182 simple reversible color transform into the processing
183 which significantly improves the compression.
184 Use this switch in combination with the -block N
185 switch (see above) for lossless JPEG coding.
186 CAUTION: A decoder with inverse color transform
187 support is required for this feature. Reversible
188 color transform support is not yet widely implemented,
189 so many decoders will be unable to view a reversible
190 color transformed JPEG file at all.
191
192 -bgycc Create big gamut YCC JPEG file.
193 In this type of encoding the color difference
194 components are quantized further by a factor of 2
195 compared to the normal Cb/Cr values, thus creating
196 space to allow larger color values with higher
197 saturation than the normal gamut limits to be encoded.
198 In order to compensate for the loss of color fidelity
199 compared to a normal YCC encoded file, the color
200 quantization tables can be adjusted accordingly.
201 For example, cjpeg -bgycc -quality 80,90 will give
202 similar results as cjpeg -quality 80.
203 CAUTION: For correct decompression a decoder with big
204 gamut YCC support (JFIF version 2) is required.
205 An old decoder may or may not display a big gamut YCC
206 encoded JPEG file, depending on JFIF version check
207 and corresponding warning/error configuration.
208 In case of a granted decompression the old decoder
209 will display the image with half saturated colors.
210
211 -dct int Use integer DCT method (default).
212 -dct fast Use fast integer DCT (less accurate).
213 -dct float Use floating-point DCT method.
214 The float method is very slightly more accurate than
215 the int method, but is much slower unless your machine
216 has very fast floating-point hardware. Also note that
217 results of the floating-point method may vary slightly
218 across machines, while the integer methods should give
219 the same results everywhere. The fast integer method
220 is much less accurate than the other two.
221
222 -nosmooth Don't use high-quality downsampling.
223
224 -restart N Emit a JPEG restart marker every N MCU rows, or every
225 N MCU blocks if "B" is attached to the number.
226 -restart 0 (the default) means no restart markers.
227
228 -smooth N Smooth the input image to eliminate dithering noise.
229 N, ranging from 1 to 100, indicates the strength of
230 smoothing. 0 (the default) means no smoothing.
231
232 -maxmemory N Set limit for amount of memory to use in processing
233 large images. Value is in thousands of bytes, or
234 millions of bytes if "M" is attached to the number.
235 For example, -max 4m selects 4000000 bytes. If more
236 space is needed, temporary files will be used.
237
238 -verbose Enable debug printout. More -v's give more printout.
239 or -debug Also, version information is printed at startup.
240
241 The -restart option inserts extra markers that allow a JPEG decoder to
242 resynchronize after a transmission error. Without restart markers, any damage
243 to a compressed file will usually ruin the image from the point of the error
244 to the end of the image; with restart markers, the damage is usually confined
245 to the portion of the image up to the next restart marker. Of course, the
246 restart markers occupy extra space. We recommend -restart 1 for images that
247 will be transmitted across unreliable networks such as Usenet.
248
249 The -smooth option filters the input to eliminate fine-scale noise. This is
250 often useful when converting dithered images to JPEG: a moderate smoothing
251 factor of 10 to 50 gets rid of dithering patterns in the input file, resulting
252 in a smaller JPEG file and a better-looking image. Too large a smoothing
253 factor will visibly blur the image, however.
254
255 Switches for wizards:
256
257 -baseline Force baseline-compatible quantization tables to be
258 generated. This clamps quantization values to 8 bits
259 even at low quality settings. (This switch is poorly
260 named, since it does not ensure that the output is
261 actually baseline JPEG. For example, you can use
262 -baseline and -progressive together.)
263
264 -qtables file Use the quantization tables given in the specified
265 text file.
266
267 -qslots N[,...] Select which quantization table to use for each color
268 component.
269
270 -sample HxV[,...] Set JPEG sampling factors for each color component.
271
272 -scans file Use the scan script given in the specified text file.
273
274 The "wizard" switches are intended for experimentation with JPEG. If you
275 don't know what you are doing, DON'T USE THEM. These switches are documented
276 further in the file wizard.txt.
277
278
279 DJPEG DETAILS
280
281 The basic command line switches for djpeg are:
282
283 -colors N Reduce image to at most N colors. This reduces the
284 or -quantize N number of colors used in the output image, so that it
285 can be displayed on a colormapped display or stored in
286 a colormapped file format. For example, if you have
287 an 8-bit display, you'd need to reduce to 256 or fewer
288 colors. (-colors is the recommended name, -quantize
289 is provided only for backwards compatibility.)
290
291 -fast Select recommended processing options for fast, low
292 quality output. (The default options are chosen for
293 highest quality output.) Currently, this is equivalent
294 to "-dct fast -nosmooth -onepass -dither ordered".
295
296 -grayscale Force grayscale output even if JPEG file is color.
297 Useful for viewing on monochrome displays; also,
298 djpeg runs noticeably faster in this mode.
299
300 -rgb Force RGB output even if JPEG file is grayscale.
301 This is provided to support applications that don't
302 want to cope with grayscale as a separate case.
303
304 -scale M/N Scale the output image by a factor M/N. Currently
305 supported scale factors are M/N with all M from 1 to
306 16, where N is the source DCT size, which is 8 for
307 baseline JPEG. If the /N part is omitted, then M
308 specifies the DCT scaled size to be applied on the
309 given input. For baseline JPEG this is equivalent to
310 M/8 scaling, since the source DCT size for baseline
311 JPEG is 8. Scaling is handy if the image is larger
312 than your screen; also, djpeg runs much faster when
313 scaling down the output.
314
315 -bmp Select BMP output format (Windows flavor). 8-bit
316 colormapped format is emitted if -colors or -grayscale
317 is specified, or if the JPEG file is grayscale;
318 otherwise, 24-bit full-color format is emitted.
319
320 -gif Select GIF output format (LZW compressed).
321 Since GIF does not support more than 256 colors,
322 -colors 256 is assumed (unless you specify a smaller
323 number of colors). If you specify -fast, the default
324 number of colors is 216.
325
326 -gif0 Select GIF output format (uncompressed).
327 Since GIF does not support more than 256 colors,
328 -colors 256 is assumed (unless you specify a smaller
329 number of colors). If you specify -fast, the default
330 number of colors is 216.
331
332 -os2 Select BMP output format (OS/2 1.x flavor). 8-bit
333 colormapped format is emitted if -colors or -grayscale
334 is specified, or if the JPEG file is grayscale;
335 otherwise, 24-bit full-color format is emitted.
336
337 -pnm Select PBMPLUS (PPM/PGM) output format (this is the
338 default format). PGM is emitted if the JPEG file is
339 grayscale or if -grayscale is specified; otherwise
340 PPM is emitted.
341
342 -rle Select RLE output format. (Requires URT library.)
343
344 -targa Select Targa output format. Grayscale format is
345 emitted if the JPEG file is grayscale or if
346 -grayscale is specified; otherwise, colormapped format
347 is emitted if -colors is specified; otherwise, 24-bit
348 full-color format is emitted.
349
350 Switches for advanced users:
351
352 -dct int Use integer DCT method (default).
353 -dct fast Use fast integer DCT (less accurate).
354 -dct float Use floating-point DCT method.
355 The float method is very slightly more accurate than
356 the int method, but is much slower unless your machine
357 has very fast floating-point hardware. Also note that
358 results of the floating-point method may vary slightly
359 across machines, while the integer methods should give
360 the same results everywhere. The fast integer method
361 is much less accurate than the other two.
362
363 -dither fs Use Floyd-Steinberg dithering in color quantization.
364 -dither ordered Use ordered dithering in color quantization.
365 -dither none Do not use dithering in color quantization.
366 By default, Floyd-Steinberg dithering is applied when
367 quantizing colors; this is slow but usually produces
368 the best results. Ordered dither is a compromise
369 between speed and quality; no dithering is fast but
370 usually looks awful. Note that these switches have
371 no effect unless color quantization is being done.
372 Ordered dither is only available in -onepass mode.
373
374 -map FILE Quantize to the colors used in the specified image
375 file. This is useful for producing multiple files
376 with identical color maps, or for forcing a predefined
377 set of colors to be used. The FILE must be a GIF
378 or PPM file. This option overrides -colors and
379 -onepass.
380
381 -nosmooth Don't use high-quality upsampling.
382
383 -onepass Use one-pass instead of two-pass color quantization.
384 The one-pass method is faster and needs less memory,
385 but it produces a lower-quality image. -onepass is
386 ignored unless you also say -colors N. Also,
387 the one-pass method is always used for grayscale
388 output (the two-pass method is no improvement then).
389
390 -maxmemory N Set limit for amount of memory to use in processing
391 large images. Value is in thousands of bytes, or
392 millions of bytes if "M" is attached to the number.
393 For example, -max 4m selects 4000000 bytes. If more
394 space is needed, temporary files will be used.
395
396 -verbose Enable debug printout. More -v's give more printout.
397 or -debug Also, version information is printed at startup.
398
399
400 HINTS FOR CJPEG
401
402 Color GIF files are not the ideal input for JPEG; JPEG is really intended for
403 compressing full-color (24-bit) images. In particular, don't try to convert
404 cartoons, line drawings, and other images that have only a few distinct
405 colors. GIF works great on these, JPEG does not. If you want to convert a
406 GIF to JPEG, you should experiment with cjpeg's -quality and -smooth options
407 to get a satisfactory conversion. -smooth 10 or so is often helpful.
408
409 Avoid running an image through a series of JPEG compression/decompression
410 cycles. Image quality loss will accumulate; after ten or so cycles the image
411 may be noticeably worse than it was after one cycle. It's best to use a
412 lossless format while manipulating an image, then convert to JPEG format when
413 you are ready to file the image away.
414
415 The -optimize option to cjpeg is worth using when you are making a "final"
416 version for posting or archiving. It's also a win when you are using low
417 quality settings to make very small JPEG files; the percentage improvement
418 is often a lot more than it is on larger files. (At present, -optimize
419 mode is always selected when generating progressive JPEG files.)
420
421
422 HINTS FOR DJPEG
423
424 To get a quick preview of an image, use the -grayscale and/or -scale switches.
425 "-grayscale -scale 1/8" is the fastest case.
426
427 Several options are available that trade off image quality to gain speed.
428 "-fast" turns on the recommended settings.
429
430 "-dct fast" and/or "-nosmooth" gain speed at a small sacrifice in quality.
431 When producing a color-quantized image, "-onepass -dither ordered" is fast but
432 much lower quality than the default behavior. "-dither none" may give
433 acceptable results in two-pass mode, but is seldom tolerable in one-pass mode.
434
435 If you are fortunate enough to have very fast floating point hardware,
436 "-dct float" may be even faster than "-dct fast". But on most machines
437 "-dct float" is slower than "-dct int"; in this case it is not worth using,
438 because its theoretical accuracy advantage is too small to be significant
439 in practice.
440
441 Two-pass color quantization requires a good deal of memory; on MS-DOS machines
442 it may run out of memory even with -maxmemory 0. In that case you can still
443 decompress, with some loss of image quality, by specifying -onepass for
444 one-pass quantization.
445
446
447 HINTS FOR BOTH PROGRAMS
448
449 If more space is needed than will fit in the available main memory (as
450 determined by -maxmemory), temporary files will be used. (MS-DOS versions
451 will try to get extended or expanded memory first.) The temporary files are
452 often rather large: in typical cases they occupy three bytes per pixel, for
453 example 3*800*600 = 1.44Mb for an 800x600 image. If you don't have enough
454 free disk space, leave out -progressive and -optimize (for cjpeg) or specify
455 -onepass (for djpeg).
456
457 On MS-DOS, the temporary files are created in the directory named by the TMP
458 or TEMP environment variable, or in the current directory if neither of those
459 exist. Amiga implementations put the temp files in the directory named by
460 JPEGTMP:, so be sure to assign JPEGTMP: to a disk partition with adequate free
461 space.
462
463 The default memory usage limit (-maxmemory) is set when the software is
464 compiled. If you get an "insufficient memory" error, try specifying a smaller
465 -maxmemory value, even -maxmemory 0 to use the absolute minimum space. You
466 may want to recompile with a smaller default value if this happens often.
467
468 On machines that have "environment" variables, you can define the environment
469 variable JPEGMEM to set the default memory limit. The value is specified as
470 described for the -maxmemory switch. JPEGMEM overrides the default value
471 specified when the program was compiled, and itself is overridden by an
472 explicit -maxmemory switch.
473
474 On MS-DOS machines, -maxmemory is the amount of main (conventional) memory to
475 use. (Extended or expanded memory is also used if available.) Most
476 DOS-specific versions of this software do their own memory space estimation
477 and do not need you to specify -maxmemory.
478
479
480 JPEGTRAN
481
482 jpegtran performs various useful transformations of JPEG files.
483 It can translate the coded representation from one variant of JPEG to another,
484 for example from baseline JPEG to progressive JPEG or vice versa. It can also
485 perform some rearrangements of the image data, for example turning an image
486 from landscape to portrait format by rotation. For EXIF files and JPEG files
487 containing Exif data, you may prefer to use exiftran instead.
488
489 jpegtran works by rearranging the compressed data (DCT coefficients), without
490 ever fully decoding the image. Therefore, its transformations are lossless:
491 there is no image degradation at all, which would not be true if you used
492 djpeg followed by cjpeg to accomplish the same conversion. But by the same
493 token, jpegtran cannot perform lossy operations such as changing the image
494 quality. However, while the image data is losslessly transformed, metadata
495 can be removed. See the -copy option for specifics.
496
497 jpegtran uses a command line syntax similar to cjpeg or djpeg.
498 On Unix-like systems, you say:
499 jpegtran [switches] [inputfile] >outputfile
500 On most non-Unix systems, you say:
501 jpegtran [switches] inputfile outputfile
502 where both the input and output files are JPEG files.
503
504 To specify the coded JPEG representation used in the output file,
505 jpegtran accepts a subset of the switches recognized by cjpeg:
506 -optimize Perform optimization of entropy encoding parameters.
507 -progressive Create progressive JPEG file.
508 -arithmetic Use arithmetic coding.
509 -restart N Emit a JPEG restart marker every N MCU rows, or every
510 N MCU blocks if "B" is attached to the number.
511 -scans file Use the scan script given in the specified text file.
512 See the previous discussion of cjpeg for more details about these switches.
513 If you specify none of these switches, you get a plain baseline-JPEG output
514 file. The quality setting and so forth are determined by the input file.
515
516 The image can be losslessly transformed by giving one of these switches:
517 -flip horizontal Mirror image horizontally (left-right).
518 -flip vertical Mirror image vertically (top-bottom).
519 -rotate 90 Rotate image 90 degrees clockwise.
520 -rotate 180 Rotate image 180 degrees.
521 -rotate 270 Rotate image 270 degrees clockwise (or 90 ccw).
522 -transpose Transpose image (across UL-to-LR axis).
523 -transverse Transverse transpose (across UR-to-LL axis).
524
525 The transpose transformation has no restrictions regarding image dimensions.
526 The other transformations operate rather oddly if the image dimensions are not
527 a multiple of the iMCU size (usually 8 or 16 pixels), because they can only
528 transform complete blocks of DCT coefficient data in the desired way.
529
530 jpegtran's default behavior when transforming an odd-size image is designed
531 to preserve exact reversibility and mathematical consistency of the
532 transformation set. As stated, transpose is able to flip the entire image
533 area. Horizontal mirroring leaves any partial iMCU column at the right edge
534 untouched, but is able to flip all rows of the image. Similarly, vertical
535 mirroring leaves any partial iMCU row at the bottom edge untouched, but is
536 able to flip all columns. The other transforms can be built up as sequences
537 of transpose and flip operations; for consistency, their actions on edge
538 pixels are defined to be the same as the end result of the corresponding
539 transpose-and-flip sequence.
540
541 For practical use, you may prefer to discard any untransformable edge pixels
542 rather than having a strange-looking strip along the right and/or bottom edges
543 of a transformed image. To do this, add the -trim switch:
544 -trim Drop non-transformable edge blocks.
545 Obviously, a transformation with -trim is not reversible, so strictly speaking
546 jpegtran with this switch is not lossless. Also, the expected mathematical
547 equivalences between the transformations no longer hold. For example,
548 "-rot 270 -trim" trims only the bottom edge, but "-rot 90 -trim" followed by
549 "-rot 180 -trim" trims both edges.
550
551 If you are only interested in perfect transformation, add the -perfect switch:
552 -perfect Fails with an error if the transformation is not
553 perfect.
554 For example you may want to do
555 jpegtran -rot 90 -perfect foo.jpg || djpeg foo.jpg | pnmflip -r90 | cjpeg
556 to do a perfect rotation if available or an approximated one if not.
557
558 We also offer a lossless-crop option, which discards data outside a given
559 image region but losslessly preserves what is inside. Like the rotate and
560 flip transforms, lossless crop is restricted by the current JPEG format: the
561 upper left corner of the selected region must fall on an iMCU boundary. If
562 this does not hold for the given crop parameters, we silently move the upper
563 left corner up and/or left to make it so, simultaneously increasing the
564 region dimensions to keep the lower right crop corner unchanged. (Thus, the
565 output image covers at least the requested region, but may cover more.)
566 The adjustment of the region dimensions may be optionally disabled by
567 attaching an 'f' character ("force") to the width or height number.
568
569 The image can be losslessly cropped by giving the switch:
570 -crop WxH+X+Y Crop to a rectangular subarea of width W, height H
571 starting at point X,Y.
572
573 Crop extension: The width or height parameters can be made larger than the
574 source image. In this case the extra area is filled in with zero (neutral
575 gray). A larger width parameter has two more options: Attaching an 'f'
576 character ("flatten") to the width number will fill in the extra area with
577 the DC of the adjacent block, instead of gray out. Attaching an 'r'
578 character ("reflect") to the width number will fill in the extra area with
579 repeated reflections of the source region, instead of gray out.
580
581 A complementary lossless-wipe option is provided to discard (gray out) data
582 inside a given image region while losslessly preserving what is outside:
583 -wipe WxH+X+Y Wipe (gray out) a rectangular subarea of
584 width W, height H starting at point X,Y.
585
586 Attaching an 'f' character ("flatten") to the width number will fill the
587 region with the average of adjacent blocks, instead of gray out. In case
588 the wipe region and outside area form two horizontally adjacent rectangles,
589 attaching an 'r' character ("reflect") to the width number will fill the
590 region with repeated reflections of the outside area, instead of gray out.
591
592 Another option is lossless-drop, which replaces data at a given image
593 position by another image:
594 -drop +X+Y filename Drop another image
595
596 Both source images must have the same subsampling values. It is best if
597 they also have the same quantization, otherwise quantization adaption occurs.
598 The trim option can be used with the drop option to requantize the drop file
599 to the source file.
600
601 Other not-strictly-lossless transformation switches are:
602
603 -grayscale Force grayscale output.
604 This option discards the chrominance channels if the input image is YCbCr
605 (ie, a standard color JPEG), resulting in a grayscale JPEG file. The
606 luminance channel is preserved exactly, so this is a better method of reducing
607 to grayscale than decompression, conversion, and recompression. This switch
608 is particularly handy for fixing a monochrome picture that was mistakenly
609 encoded as a color JPEG. (In such a case, the space savings from getting rid
610 of the near-empty chroma channels won't be large; but the decoding time for
611 a grayscale JPEG is substantially less than that for a color JPEG.)
612
613 -scale M/N Scale the output image by a factor M/N.
614 Currently supported scale factors are M/N with all M from 1 to 16, where N is
615 the source DCT size, which is 8 for baseline JPEG. If the /N part is omitted,
616 then M specifies the DCT scaled size to be applied on the given input. For
617 baseline JPEG this is equivalent to M/8 scaling, since the source DCT size
618 for baseline JPEG is 8. CAUTION: An implementation of the JPEG SmartScale
619 extension is required for this feature. SmartScale enabled JPEG is not yet
620 widely implemented, so many decoders will be unable to view a SmartScale
621 extended JPEG file at all.
622
623 jpegtran also recognizes these switches that control what to do with "extra"
624 markers, such as comment blocks:
625 -copy none Copy no extra markers from source file.
626 This setting suppresses all comments
627 and other metadata in the source file.
628 -copy comments Copy only comment markers.
629 This setting copies comments from the source file,
630 but discards any other metadata.
631 -copy all Copy all extra markers. This setting preserves
632 metadata found in the source file, such as JFIF
633 thumbnails, Exif data, and Photoshop settings.
634 In some files these extra markers can be sizable.
635 Note that this option will copy thumbnails as-is;
636 they will not be transformed.
637 The default behavior is -copy comments. (Note: in IJG releases v6 and v6a,
638 jpegtran always did the equivalent of -copy none.)
639
640 Additional switches recognized by jpegtran are:
641 -outfile filename
642 -maxmemory N
643 -verbose
644 -debug
645 These work the same as in cjpeg or djpeg.
646
647
648 THE COMMENT UTILITIES
649
650 The JPEG standard allows "comment" (COM) blocks to occur within a JPEG file.
651 Although the standard doesn't actually define what COM blocks are for, they
652 are widely used to hold user-supplied text strings. This lets you add
653 annotations, titles, index terms, etc to your JPEG files, and later retrieve
654 them as text. COM blocks do not interfere with the image stored in the JPEG
655 file. The maximum size of a COM block is 64K, but you can have as many of
656 them as you like in one JPEG file.
657
658 We provide two utility programs to display COM block contents and add COM
659 blocks to a JPEG file.
660
661 rdjpgcom searches a JPEG file and prints the contents of any COM blocks on
662 standard output. The command line syntax is
663 rdjpgcom [-raw] [-verbose] [inputfilename]
664 The switch "-raw" (or just "-r") causes rdjpgcom to also output non-printable
665 characters in comments, which are normally escaped for security reasons.
666 The switch "-verbose" (or just "-v") causes rdjpgcom to also display the JPEG
667 image dimensions. If you omit the input file name from the command line,
668 the JPEG file is read from standard input. (This may not work on some
669 operating systems, if binary data can't be read from stdin.)
670
671 wrjpgcom adds a COM block, containing text you provide, to a JPEG file.
672 Ordinarily, the COM block is added after any existing COM blocks, but you
673 can delete the old COM blocks if you wish. wrjpgcom produces a new JPEG
674 file; it does not modify the input file. DO NOT try to overwrite the input
675 file by directing wrjpgcom's output back into it; on most systems this will
676 just destroy your file.
677
678 The command line syntax for wrjpgcom is similar to cjpeg's. On Unix-like
679 systems, it is
680 wrjpgcom [switches] [inputfilename]
681 The output file is written to standard output. The input file comes from
682 the named file, or from standard input if no input file is named.
683
684 On most non-Unix systems, the syntax is
685 wrjpgcom [switches] inputfilename outputfilename
686 where both input and output file names must be given explicitly.
687
688 wrjpgcom understands three switches:
689 -replace Delete any existing COM blocks from the file.
690 -comment "Comment text" Supply new COM text on command line.
691 -cfile name Read text for new COM block from named file.
692 (Switch names can be abbreviated.) If you have only one line of comment text
693 to add, you can provide it on the command line with -comment. The comment
694 text must be surrounded with quotes so that it is treated as a single
695 argument. Longer comments can be read from a text file.
696
697 If you give neither -comment nor -cfile, then wrjpgcom will read the comment
698 text from standard input. (In this case an input image file name MUST be
699 supplied, so that the source JPEG file comes from somewhere else.) You can
700 enter multiple lines, up to 64KB worth. Type an end-of-file indicator
701 (usually control-D or control-Z) to terminate the comment text entry.
702
703 wrjpgcom will not add a COM block if the provided comment string is empty.
704 Therefore -replace -comment "" can be used to delete all COM blocks from a
705 file.
706
707 These utility programs do not depend on the IJG JPEG library. In
708 particular, the source code for rdjpgcom is intended as an illustration of
709 the minimum amount of code required to parse a JPEG file header correctly.