1 .TH JPEGTRAN 1 "3 August 1997"
3 jpegtran \- lossless transformation of JPEG files
16 performs various useful transformations of JPEG files.
17 It can translate the coded representation from one variant of JPEG to another,
18 for example from baseline JPEG to progressive JPEG or vice versa. It can also
19 perform some rearrangements of the image data, for example turning an image
20 from landscape to portrait format by rotation.
23 works by rearranging the compressed data (DCT coefficients), without
24 ever fully decoding the image. Therefore, its transformations are lossless:
25 there is no image degradation at all, which would not be true if you used
29 to accomplish the same conversion. But by the same token,
31 cannot perform lossy operations such as changing the image quality.
34 reads the named JPEG/JFIF file, or the standard input if no file is
35 named, and produces a JPEG/JFIF file on the standard output.
37 All switch names may be abbreviated; for example,
43 Upper and lower case are equivalent.
44 British spellings are also accepted (e.g.,
46 though for brevity these are not mentioned below.
48 To specify the coded JPEG representation used in the output file,
50 accepts a subset of the switches recognized by
54 Perform optimization of entropy encoding parameters.
57 Create progressive JPEG file.
60 Emit a JPEG restart marker every N MCU rows, or every N MCU blocks if "B" is
61 attached to the number.
64 Use the scan script given in the specified text file.
68 for more details about these switches.
69 If you specify none of these switches, you get a plain baseline-JPEG output
70 file. The quality setting and so forth are determined by the input file.
72 The image can be losslessly transformed by giving one of these switches:
75 Mirror image horizontally (left-right).
78 Mirror image vertically (top-bottom).
81 Rotate image 90 degrees clockwise.
84 Rotate image 180 degrees.
87 Rotate image 270 degrees clockwise (or 90 ccw).
90 Transpose image (across UL-to-LR axis).
93 Transverse transpose (across UR-to-LL axis).
95 The transpose transformation has no restrictions regarding image dimensions.
96 The other transformations operate rather oddly if the image dimensions are not
97 a multiple of the iMCU size (usually 8 or 16 pixels), because they can only
98 transform complete blocks of DCT coefficient data in the desired way.
101 default behavior when transforming an odd-size image is designed
102 to preserve exact reversibility and mathematical consistency of the
103 transformation set. As stated, transpose is able to flip the entire image
104 area. Horizontal mirroring leaves any partial iMCU column at the right edge
105 untouched, but is able to flip all rows of the image. Similarly, vertical
106 mirroring leaves any partial iMCU row at the bottom edge untouched, but is
107 able to flip all columns. The other transforms can be built up as sequences
108 of transpose and flip operations; for consistency, their actions on edge
109 pixels are defined to be the same as the end result of the corresponding
110 transpose-and-flip sequence.
112 For practical use, you may prefer to discard any untransformable edge pixels
113 rather than having a strange-looking strip along the right and/or bottom edges
114 of a transformed image. To do this, add the
119 Drop non-transformable edge blocks.
121 Obviously, a transformation with
123 is not reversible, so strictly speaking
125 with this switch is not lossless. Also, the expected mathematical
126 equivalences between the transformations no longer hold. For example,
128 trims only the bottom edge, but
134 Another not-strictly-lossless transformation switch is:
137 Force grayscale output.
139 This option discards the chrominance channels if the input image is YCbCr
140 (ie, a standard color JPEG), resulting in a grayscale JPEG file. The
141 luminance channel is preserved exactly, so this is a better method of reducing
142 to grayscale than decompression, conversion, and recompression. This switch
143 is particularly handy for fixing a monochrome picture that was mistakenly
144 encoded as a color JPEG. (In such a case, the space savings from getting rid
145 of the near-empty chroma channels won't be large; but the decoding time for
146 a grayscale JPEG is substantially less than that for a color JPEG.)
149 also recognizes these switches that control what to do with "extra" markers,
150 such as comment blocks:
153 Copy no extra markers from source file. This setting suppresses all
154 comments and other excess baggage present in the source file.
157 Copy only comment markers. This setting copies comments from the source file,
158 but discards any other inessential data.
161 Copy all extra markers. This setting preserves miscellaneous markers
162 found in the source file, such as JFIF thumbnails and Photoshop settings.
163 In some files these extra markers can be sizable.
165 The default behavior is
166 .BR "\-copy comments" .
167 (Note: in IJG releases v6 and v6a,
169 always did the equivalent of
172 Additional switches recognized by jpegtran are:
175 Set limit for amount of memory to use in processing large images. Value is
176 in thousands of bytes, or millions of bytes if "M" is attached to the
179 selects 4000000 bytes. If more space is needed, temporary files will be used.
181 .BI \-outfile " name"
182 Send output image to the named file, not to standard output.
185 Enable debug printout. More
187 give more output. Also, version information is printed at startup.
194 This example converts a baseline JPEG file to progressive form:
196 .B jpegtran \-progressive
201 This example rotates an image 90 degrees clockwise, discarding any
202 unrotatable edge pixels:
204 .B jpegtran \-rot 90 -trim
211 If this environment variable is set, its value is the default memory limit.
212 The value is specified as described for the
216 overrides the default value specified when the program was compiled, and
217 itself is overridden by an explicit
225 Wallace, Gregory K. "The JPEG Still Picture Compression Standard",
226 Communications of the ACM, April 1991 (vol. 34, no. 4), pp. 30-44.
228 Independent JPEG Group
230 Arithmetic coding is not supported for legal reasons.
232 The transform options can't transform odd-size images perfectly. Use
234 if you don't like the results without it.
236 The entire image is read into memory and then written out again, even in
237 cases where this isn't really necessary. Expect swapping on large images,
238 especially when using the more complex transform options.