官方教程

https://github.com/mozilla/mozjpeg/blob/master/BUILDING.txt

源码地址

https://github.com/mozilla/mozjpeg/releases

需 nasm 环境

yum -y install build-essential nasm

以 3.3.1 为例

cd ~
wget -O "mozjpeg-3.3.1.tar.gz" https://codeload.github.com/mozilla/mozjpeg/tar.gz/v3.3.1
tar -zxvf mozjpeg-v3.3.1.tar.gz 
cd mozjpeg-3.3.1
autoreconf -fiv
./configure
make
make install

安装完成 目录

/opt/mozjpeg/bin
-rwxr-xr-x 1 root root 56752 Nov  2 23:26 cjpeg
-rwxr-xr-x 1 root root 45792 Nov  2 23:26 djpeg
-rwxr-xr-x 1 root root 53424 Nov  2 23:26 jpegtran
-rwxr-xr-x 1 root root 13536 Nov  2 23:26 rdjpgcom
-rwxr-xr-x 1 root root 64456 Nov  2 23:26 tjbench
-rwxr-xr-x 1 root root 13552 Nov  2 23:26 wrjpgcom

使用示例

/opt/mozjpeg/bin/cjpeg -quality 80 a.jpg > a1.jpg

图片由 4.9MB 压缩至 1.4MB

帮助文档

[root@instance-jewlel2q bin]# ./cjpeg -h
./cjpeg: unknown option 'h'
usage: ./cjpeg [switches] [inputfile]
Switches (names may be abbreviated):
  -quality N[,...]   Compression quality (0..100; 5-95 is most useful range,
                     default is 75)
  -grayscale     Create monochrome JPEG file
  -rgb           Create RGB JPEG file
  -optimize      Optimize Huffman table (smaller file, but slow compression, enabled by default)
  -progressive   Create progressive JPEG file (enabled by default)
  -baseline      Create baseline JPEG file (disable progressive coding)
  -targa         Input file is Targa format (usually not needed)
  -revert        Revert to standard defaults (instead of mozjpeg defaults)
  -fastcrush     Disable progressive scan optimization
  -dc-scan-opt   DC scan optimization mode
                 - 0 One scan for all components
                 - 1 One scan per component (default)
                 - 2 Optimize between one scan for all components and one scan for 1st component
                     plus one scan for remaining components
  -notrellis     Disable trellis optimization
  -trellis-dc    Enable trellis optimization of DC coefficients (default)
  -notrellis-dc  Disable trellis optimization of DC coefficients
  -tune-psnr     Tune trellis optimization for PSNR
  -tune-hvs-psnr Tune trellis optimization for PSNR-HVS (default)
  -tune-ssim     Tune trellis optimization for SSIM
  -tune-ms-ssim  Tune trellis optimization for MS-SSIM
Switches for advanced users:
  -noovershoot   Disable black-on-white deringing via overshoot
  -arithmetic    Use arithmetic coding
  -dct int       Use integer DCT method (default)
  -dct fast      Use fast integer DCT (less accurate)
  -dct float     Use floating-point DCT method
  -quant-baseline Use 8-bit quantization table entries for baseline JPEG compatibility
  -quant-table N Use predefined quantization table N:
                 - 0 JPEG Annex K
                 - 1 Flat
                 - 2 Custom, tuned for MS-SSIM
                 - 3 ImageMagick table by N. Robidoux
                 - 4 Custom, tuned for PSNR-HVS
                 - 5 Table from paper by Klein, Silverstein and Carney
  -restart N     Set restart interval in rows, or in blocks with B
  -smooth N      Smooth dithered input (N=1..100 is strength)
  -maxmemory N   Maximum memory to use (in kbytes)
  -outfile name  Specify name for output file
  -memdst        Compress to memory instead of file (useful for benchmarking)
  -verbose  or  -debug   Emit debug output
  -version       Print version information and exit
Switches for wizards:
  -qtables FILE  Use quantization tables given in FILE
  -qslots N[,...]    Set component quantization tables
  -sample HxV[,...]  Set component sampling factors
  -scans FILE    Create multi-scan JPEG per script FILE
[root@instance-jewlel2q bin]# ./djpeg -h
usage: ./djpeg [switches] [inputfile]
Switches (names may be abbreviated):
  -colors N      Reduce image to no more than N colors
  -fast          Fast, low-quality processing
  -grayscale     Force grayscale output
  -rgb           Force RGB output
  -rgb565        Force RGB565 output
  -scale M/N     Scale output image by fraction M/N, eg, 1/8
  -bmp           Select BMP output format (Windows style)
  -gif           Select GIF output format
  -os2           Select BMP output format (OS/2 style)
  -pnm           Select PBMPLUS (PPM/PGM) output format (default)
  -targa         Select Targa output format
Switches for advanced users:
  -dct int       Use integer DCT method (default)
  -dct fast      Use fast integer DCT (less accurate)
  -dct float     Use floating-point DCT method
  -dither fs     Use F-S dithering (default)
  -dither none   Don't use dithering in quantization
  -dither ordered  Use ordered dither (medium speed, quality)
  -map FILE      Map to colors used in named image file
  -nosmooth      Don't use high-quality upsampling
  -onepass       Use 1-pass quantization (fast, low quality)
  -maxmemory N   Maximum memory to use (in kbytes)
  -outfile name  Specify name for output file
  -memsrc        Load input file into memory before decompressing
  -skip Y0,Y1    Decompress all rows except those between Y0 and Y1 (inclusive)
  -crop WxH+X+Y  Decompress only a rectangular subregion of the image
  -verbose  or  -debug   Emit debug output
  -version       Print version information and exit
[root@instance-jewlel2q bin]# ./jpegtran -h
usage: ./jpegtran [switches] [inputfile]
Switches (names may be abbreviated):
  -copy none     Copy no extra markers from source file
  -copy comments Copy only comment markers (default)
  -copy all      Copy all extra markers
  -optimize      Optimize Huffman table (smaller file, but slow compression, enabled by default)
  -progressive   Create progressive JPEG file (enabled by default)
  -revert        Revert to standard defaults (instead of mozjpeg defaults)
  -fastcrush     Disable progressive scan optimization
Switches for modifying the image:
  -crop WxH+X+Y  Crop to a rectangular subarea
  -grayscale     Reduce to grayscale (omit color data)
  -flip [horizontal|vertical]  Mirror image (left-right or top-bottom)
  -perfect       Fail if there is non-transformable edge blocks
  -rotate [90|180|270]         Rotate image (degrees clockwise)
  -transpose     Transpose image
  -transverse    Transverse transpose image
  -trim          Drop non-transformable edge blocks
Switches for advanced users:
  -arithmetic    Use arithmetic coding
  -restart N     Set restart interval in rows, or in blocks with B
  -maxmemory N   Maximum memory to use (in kbytes)
  -outfile name  Specify name for output file
  -verbose  or  -debug   Emit debug output
  -version       Print version information and exit
Switches for wizards:
  -scans FILE    Create multi-scan JPEG per script FILE
[root@instance-jewlel2q bin]# ./rdjpgcom -h
rdjpgcom displays any textual comments in a JPEG file.
Usage: ./rdjpgcom [switches] [inputfile]
Switches (names may be abbreviated):
  -raw        Display non-printable characters in comments (unsafe)
  -verbose    Also display dimensions of JPEG image
[root@instance-jewlel2q bin]# ./tjbench -h

USAGE: ./tjbench
       <Inputfile (BMP|PPM)> <Quality> [options]

       ./tjbench
       <Inputfile (JPG)> [options]

Options:

-alloc = Dynamically allocate JPEG image buffers
-bmp = Generate output images in Windows Bitmap format (default = PPM)
-bottomup = Test bottom-up compression/decompression
-tile = Test performance of the codec when the image is encoded as separate
     tiles of varying sizes.
-rgb, -bgr, -rgbx, -bgrx, -xbgr, -xrgb =
     Test the specified color conversion path in the codec (default = BGR)
-cmyk = Indirectly test YCCK JPEG compression/decompression (the source
     and destination bitmaps are still RGB.  The conversion is done
     internally prior to compression or after decompression.)
-fastupsample = Use the fastest chrominance upsampling algorithm available in
     the underlying codec
-fastdct = Use the fastest DCT/IDCT algorithms available in the underlying
     codec
-accuratedct = Use the most accurate DCT/IDCT algorithms available in the
     underlying codec
-subsamp <s> = When testing JPEG compression, this option specifies the level
     of chrominance subsampling to use (<s> = 444, 422, 440, 420, 411, or
     GRAY).  The default is to test Grayscale, 4:2:0, 4:2:2, and 4:4:4 in
     sequence.
-quiet = Output results in tabular rather than verbose format
-yuv = Test YUV encoding/decoding functions
-yuvpad <p> = If testing YUV encoding/decoding, this specifies the number of
     bytes to which each row of each plane in the intermediate YUV image is
     padded (default = 1)
-scale M/N = Scale down the width/height of the decompressed JPEG image by a
     factor of M/N (M/N = 2/1, 15/8, 7/4, 13/8, 3/2, 11/8, 5/4, 9/8, 1/1, 
     7/8, 3/4, 5/8, 1/2, 3/8, 1/4, or 1/8)
-hflip, -vflip, -transpose, -transverse, -rot90, -rot180, -rot270 =
     Perform the corresponding lossless transform prior to
     decompression (these options are mutually exclusive)
-grayscale = Perform lossless grayscale conversion prior to decompression
     test (can be combined with the other transforms above)
-benchtime <t> = Run each benchmark for at least <t> seconds (default = 5.0)
-warmup <t> = Run each benchmark for <t> seconds (default = 1.0) prior to
     starting the timer, in order to prime the caches and thus improve the
     consistency of the results.
-componly = Stop after running compression tests.  Do not test decompression.
-nowrite = Do not write reference or output images (improves consistency of
     performance measurements.)

NOTE:  If the quality is specified as a range (e.g. 90-100), a separate
test will be performed for all quality values in the range.
[root@instance-jewlel2q bin]# ./wrjpgcom -h
wrjpgcom inserts a textual comment in a JPEG file.
You can add to or replace any existing comment(s).
Usage: ./wrjpgcom [switches] [inputfile]
Switches (names may be abbreviated):
  -replace         Delete any existing comments
  -comment "text"  Insert comment with given text
  -cfile name      Read comment from named file
Notice that you must put quotes around the comment text
when you use -comment.
If you do not give either -comment or -cfile on the command line,
then the comment text is read from standard input.
It can be multiple lines, up to 65000 characters total.
You must specify an input JPEG file name when supplying
comment text from standard input.

1 Comment

发表评论

电子邮件地址不会被公开。 必填项已用*标注