1DJPEG(1) General Commands Manual DJPEG(1)
2
6 djpeg - decompress a JPEG file to an image file
7
9 djpeg [ options ] [ filename ]
10
12 djpeg decompresses the named JPEG file, or the standard input if no
13 file is named, and produces an image file on the standard output. PBM‐
14 PLUS (PPM/PGM), BMP, GIF, Targa, or RLE (Utah Raster Toolkit) output
15 format can be selected. (RLE is supported only if the URT library is
16 available.)
17
19 All switch names may be abbreviated; for example, -grayscale may be
20 written -gray or -gr. Most of the "basic" switches can be abbreviated
21 to as little as one letter. Upper and lower case are equivalent (thus
22 -BMP is the same as -bmp). British spellings are also accepted (e.g.,
23 -greyscale), though for brevity these are not mentioned below.
24 The basic switches are:
26 -colors N
28 Reduce image to at most N colors. This reduces the number of
29 colors used in the output image, so that it can be displayed on
30 a colormapped display or stored in a colormapped file format.
31 For example, if you have an 8-bit display, you'd need to reduce
32 to 256 or fewer colors.
33 -quantize N
35 Same as -colors. -colors is the recommended name, -quantize is
36 provided only for backwards compatibility.
37 -fast Select recommended processing options for fast, low quality out‐
39 put. (The default options are chosen for highest quality out‐
40 put.) Currently, this is equivalent to -dct fast -nosmooth
41 -onepass -dither ordered.
42 -grayscale
44 Force grayscale output even if JPEG file is color. Useful for
45 viewing on monochrome displays; also, djpeg runs noticeably
46 faster in this mode.
47 -rgb Force RGB output even if JPEG file is grayscale.
49 -scale M/N
51 Scale the output image by a factor M/N. Currently the scale
52 factor must be M/8, where M is an integer between 1 and 16
53 inclusive, or any reduced fraction thereof (such as 1/2, 3/4,
54 etc.) Scaling is handy if the image is larger than your screen;
55 also, djpeg runs much faster when scaling down the output.
56 -bmp Select BMP output format (Windows flavor). 8-bit colormapped
58 format is emitted if -colors or -grayscale is specified, or if
59 the JPEG file is grayscale; otherwise, 24-bit full-color format
60 is emitted.
61 -gif Select GIF output format. Since GIF does not support more than
63 256 colors, -colors 256 is assumed (unless you specify a smaller
64 number of colors).
65 -os2 Select BMP output format (OS/2 1.x flavor). 8-bit colormapped
67 format is emitted if -colors or -grayscale is specified, or if
68 the JPEG file is grayscale; otherwise, 24-bit full-color format
69 is emitted.
70 -pnm Select PBMPLUS (PPM/PGM) output format (this is the default for‐
72 mat). PGM is emitted if the JPEG file is grayscale or if
73 -grayscale is specified; otherwise PPM is emitted.
74 -rle Select RLE output format. (Requires URT library.)
76 -targa Select Targa output format. Grayscale format is emitted if the
78 JPEG file is grayscale or if -grayscale is specified; otherwise,
79 colormapped format is emitted if -colors is specified; other‐
80 wise, 24-bit full-color format is emitted.
81 Switches for advanced users:
83 -dct int
85 Use integer DCT method (default).
86 -dct fast
88 Use fast integer DCT (less accurate). In libjpeg-turbo, the
89 fast method is generally about 5-15% faster than the int method
90 when using the x86/x86-64 SIMD extensions (results may vary with
91 other SIMD implementations, or when using libjpeg-turbo without
92 SIMD extensions.) If the JPEG image was compressed using a
93 quality level of 85 or below, then there should be little or no
94 perceptible difference between the two algorithms. When decom‐
95 pressing images that were compressed using quality levels above
96 85, however, the difference between the fast and int methods
97 becomes more pronounced. With images compressed using qual‐
98 ity=97, for instance, the fast method incurs generally about a
99 4-6 dB loss (in PSNR) relative to the int method, but this can
100 be larger for some images. If you can avoid it, do not use the
101 fast method when decompressing images that were compressed using
102 quality levels above 97. The algorithm often degenerates for
103 such images and can actually produce a more lossy output image
104 than if the JPEG image had been compressed using lower quality
105 levels.
106 -dct float
108 Use floating-point DCT method. The float method is mainly a
109 legacy feature. It does not produce significantly more accurate
110 results than the int method, and it is much slower. The float
111 method may also give different results on different machines due
112 to varying roundoff behavior, whereas the integer methods should
113 give the same results on all machines.
114 -dither fs
116 Use Floyd-Steinberg dithering in color quantization.
117 -dither ordered
119 Use ordered dithering in color quantization.
120 -dither none
122 Do not use dithering in color quantization. By default, Floyd-
123 Steinberg dithering is applied when quantizing colors; this is
124 slow but usually produces the best results. Ordered dither is a
125 compromise between speed and quality; no dithering is fast but
126 usually looks awful. Note that these switches have no effect
127 unless color quantization is being done. Ordered dither is only
128 available in -onepass mode.
129 -map file
131 Quantize to the colors used in the specified image file. This
132 is useful for producing multiple files with identical color
133 maps, or for forcing a predefined set of colors to be used. The
134 file must be a GIF or PPM file. This option overrides -colors
135 and -onepass.
136 -nosmooth
138 Use a faster, lower-quality upsampling routine.
139 -onepass
141 Use one-pass instead of two-pass color quantization. The one-
142 pass method is faster and needs less memory, but it produces a
143 lower-quality image. -onepass is ignored unless you also say
144 -colors N. Also, the one-pass method is always used for
145 grayscale output (the two-pass method is no improvement then).
146 -maxmemory N
148 Set limit for amount of memory to use in processing large
149 images. Value is in thousands of bytes, or millions of bytes if
150 "M" is attached to the number. For example, -max 4m selects
151 4000000 bytes. If more space is needed, an error will occur.
152 -outfile name
154 Send output image to the named file, not to standard output.
155 -memsrc
157 Load input file into memory before decompressing. This feature
158 was implemented mainly as a way of testing the in-memory source
159 manager (jpeg_mem_src().)
160 -skip Y0,Y1
162 Decompress all rows of the JPEG image except those between Y0
163 and Y1 (inclusive.) Note that if decompression scaling is being
164 used, then Y0 and Y1 are relative to the scaled image dimen‐
165 sions.
166 -crop WxH+X+Y
168 Decompress only a rectangular subregion of the image, starting
169 at point X,Y with width W and height H. If necessary, X will be
170 shifted left to the nearest iMCU boundary, and the width will be
171 increased accordingly. Note that if decompression scaling is
172 being used, then X, Y, W, and H are relative to the scaled image
173 dimensions. Currently this option only works with the PBMPLUS
174 (PPM/PGM), GIF, and Targa output formats.
175 -verbose
177 Enable debug printout. More -v's give more output. Also, ver‐
178 sion information is printed at startup.
179 -debug Same as -verbose.
181 -version
183 Print version information and exit.
184
186 This example decompresses the JPEG file foo.jpg, quantizes it to 256
187 colors, and saves the output in 8-bit BMP format in foo.bmp:
188 djpeg -colors 256 -bmp foo.jpg > foo.bmp
190
192 To get a quick preview of an image, use the -grayscale and/or -scale
193 switches. -grayscale -scale 1/8 is the fastest case.
194 Several options are available that trade off image quality to gain
196 speed. -fast turns on the recommended settings.
197 -dct fast and/or -nosmooth gain speed at a small sacrifice in quality.
199 When producing a color-quantized image, -onepass -dither ordered is
200 fast but much lower quality than the default behavior. -dither none
201 may give acceptable results in two-pass mode, but is seldom tolerable
202 in one-pass mode.
203 If you are fortunate enough to have very fast floating point hardware,
205 -dct float may be even faster than -dct fast. But on most machines
206 -dct float is slower than -dct int; in this case it is not worth using,
207 because its theoretical accuracy advantage is too small to be signifi‐
208 cant in practice.
209
211 JPEGMEM
212 If this environment variable is set, its value is the default
213 memory limit. The value is specified as described for the
214 -maxmemory switch. JPEGMEM overrides the default value speci‐
215 fied when the program was compiled, and itself is overridden by
216 an explicit -maxmemory.
217
219 cjpeg(1), jpegtran(1), rdjpgcom(1), wrjpgcom(1)
220 ppm(5), pgm(5)
221 Wallace, Gregory K. "The JPEG Still Picture Compression Standard",
222 Communications of the ACM, April 1991 (vol. 34, no. 4), pp. 30-44.
223
225 Independent JPEG Group
226 This file was modified by The libjpeg-turbo Project to include only
228 information relevant to libjpeg-turbo, to wordsmith certain sections,
229 and to describe features not present in libjpeg.
230
232 Support for compressed GIF output files was removed in djpeg v6b due to
233 concerns over the Unisys LZW patent. Although this patent expired in
234 2006, djpeg still lacks compressed GIF support, for these historical
235 reasons. (Conversion of JPEG files to GIF is usually a bad idea any‐
236 way, since GIF is a 256-color format.) The uncompressed GIF files that
237 djpeg generates are larger than they should be, but they are readable
238 by standard GIF decoders.
239 13 November 2017 DJPEG(1)