Provided by: netpbm_10.0-15_amd64 bug


       pnmtojpeg - convert PNM image to a JFIF ("JPEG") image


       pnmtojpeg [ options ] [ filename ]


       pnmtojpeg converts the named PBM, PGM, or PPM image file, or the standard input if no file
       is named, to a JFIF file on the standard output.

       pnmtojpeg uses the Independent JPEG Group's JPEG library to create the output  file.   See for information on the library.

       "JFIF"  is  the  correct  name  for  the  image format commonly known as "JPEG."  Strictly
       speaking, JPEG is a method of compression.  The image format using JPEG  compression  that
       is  by  far  the  most  common  is JFIF.  There is also a subformat of TIFF that uses JPEG

       EXIF is an image format that is a subformat of JFIF (to wit, a JFIF file that contains  an
       EXIF  header  as  an  APP1  marker).  pnmtojpeg creates an EXIF image when you specify the
       -exif option.


       The basic options are:

              This option specifies that the output image is to be EXIF (a  subformat  of  JFIF),
              i.e.  it  will  have  an  EXIF  header as a JFIF APP1 marker.  The contents of that
              marker are the contents of the specified file.  The special value - means  to  read
              the  EXIF  header  contents from standard input.  It is invalid to specify standard
              input for both the EXIF header and the input image.

              The EXIF file starts with a two byte  field  which  is  the  length  of  the  file,
              including  the  length  field,  in  pure  binary, most significant byte first.  The
              special value of zero for the length field means there is to  be  no  EXIF  header,
              i.e.  the same as no -exif option.  This is useful for when you convert a file from
              JFIF to PNM using jpegtopnm, then transform it, then convert it back to  JFIF  with
              pnmtojpeg, and you don't know whether or not it includes an EXIF header.  jpegtopnm
              creates an EXIF file containing nothing but two bytes of zero when the  input  JFIF
              file  has  no  EXIF  header.  Thus, you can transfer any EXIF header from the input
              JFIF to the output JFIF without worrying about  whether  an  EXIF  header  actually

              The  contents  of  the EXIF file after the length field are the exact byte for byte
              contents of the APP1 marker, not counting the length field,  that  constitutes  the
              EXIF header.

              Scale  quantization  tables to adjust image quality.  n is 0 (worst) to 100 (best);
              default is 75.  (See below for more info.)


              Create gray scale JFIF file.  With this option, pnmtojpeg converts color  input  to
              gray  scale.   If you don't specify this option, The output file is in color format
              if the input is PPM, and grayscale format if the input is PBM or PGM.

              In the PPM input case, even if all the colors in the image are gray, the output  is
              in  color  format.   Of course, the colors in it are still gray.  The difference is
              that color format takes up a lot more space and takes longer to create and process.

              Perform optimization of entropy encoding parameters.  Without this, pnmtojpeg  uses
              default  encoding  parameters.   --optimize  usually  makes  the JFIF file a little
              smaller, but pnmtojpeg runs somewhat slower and  needs  much  more  memory.   Image
              quality and speed of decompression are unaffected by --optimize.

              Create a progressive JPEG file (see below).

              Include  a comment marker in the JFIF output, with comment text text.  Without this
              option, there are no comment markers in the output.

       The --quality option lets you trade off  compressed  file  size  against  quality  of  the
       reconstructed  image:  the  higher  the quality setting, the larger the JFIF file, and the
       closer the output image will be to the original input.   Normally  you  want  to  use  the
       lowest   quality  setting  (smallest  file)  that  decompresses  into  something  visually
       indistinguishable from the original image.  For this purpose the quality setting should be
       between  50  and  95;  the  default  of  75  is  often about right.  If you see defects at
       --quality=75, then go up 5 or 10 counts at a time until you  are  happy  with  the  output
       image.  (The optimal setting will vary from one image to another.)

       --quality=100  generates  a  quantization  table  of  all  1's,  minimizing  loss  in  the
       quantization step (but there is still information loss in subsampling, as well as roundoff
       error).   This  setting  is  mainly of interest for experimental purposes.  Quality values
       above about 95 are not recommended for normal  use;  the  compressed  file  size  goes  up
       dramatically for hardly any gain in output image quality.

       In the other direction, quality values below 50 will produce very small files of low image
       quality.  Settings around 5 to 10 might be useful in preparing an index of a  large  image
       library,  for  example.   Try --quality=2 (or so) for some amusing Cubist effects.  (Note:
       quality values below about 25 generate 2-byte quantization tables,  which  are  considered
       optional  in  the  JFIF  standard.  pnmtojpeg emits a warning message when you give such a
       quality value, because some other JFIF programs may be  unable  to  decode  the  resulting
       file.  Use --baseline if you need to ensure compatibility at low quality values.)

       The  --progressive  option  creates a "progressive JPEG" file.  In this type of JFIF file,
       the data is stored in multiple  scans  of  increasing  quality.   If  the  file  is  being
       transmitted over a slow communications link, the decoder can use the first scan to display
       a low-quality image very quickly, and can then improve the display  with  each  subsequent
       scan.   The  final image is exactly equivalent to a standard JFIF file of the same quality
       setting, and the total file size is about the same -- often a  little  smaller.   Caution:
       progressive  JPEG is not yet widely implemented, so many decoders will be unable to view a
       progressive JPEG file at all.

       Options for advanced users:

              Use integer DCT method (default).

              Use fast integer DCT (less accurate).

              Use floating-point DCT method.  The float method is  very  slightly  more  accurate
              than the int method, but is much slower unless your machine has very fast floating-
              point hardware.  Also note that results  of  the  floating-point  method  may  vary
              slightly  across  machines,  while the integer methods should give the same results
              everywhere.  The fast integer method is much less accurate than the other two.

              Emit a JPEG restart marker every n MCU rows, or every n MCU blocks if you append  B
              to the number.  --restart 0 (the default) means no restart markers.

              Smooth  the  input  image  to eliminate dithering noise.  n, ranging from 1 to 100,
              indicates the strength of smoothing.  0 (the default) means no smoothing.

              Set a limit for amount of memory to use in processing large images.   Value  is  in
              thousands  of  bytes,  or  millions  of  bytes  if you append M to the number.  For
              example, --max=4m selects 4,000,000 bytes.  If pnmtojpeg needs more space, it  will
              use temporary files.

              Print  to  the Standard Error file messages about the conversion process.  This can
              be helpful in debugging problems.

       The --restart option tells pnmtojpeg to insert extra markers that allow a JPEG decoder  to
       resynchronize  after  a  transmission  error.   Without  restart  markers, any damage to a
       compressed file will usually ruin the image from the point of the error to the end of  the
       image; with restart markers, the damage is usually confined to the portion of the image up
       to the next restart marker.  Of course,  the  restart  markers  occupy  extra  space.   We
       recommend  --restart=1 for images that will be transmitted across unreliable networks such
       as Usenet.

       The --smooth option filters the input to eliminate fine-scale noise.  This is often useful
       when converting dithered images to JFIF:  a moderate smoothing factor of 10 to 50 gets rid
       of dithering patterns in the input file, resulting in a smaller JFIF file  and  a  better-
       looking image.  Too large a smoothing factor will visibly blur the image, however.

       Options for wizards:

              Force  baseline-compatible  quantization  tables  to  be  generated.   This  clamps
              quantization values to 8 bits even at low quality settings.  (This switch is poorly
              named,  since  it  does  not ensure that the output is actually baseline JPEG.  For
              example, you can use --baseline and --progressive together.)

              Use the quantization tables given in the specified text file.

              Select which quantization table to use for each color component.

              Set JPEG sampling factors for each color component.

              Use the scan script given in the specified text file.  See below for information on
              scan scripts.

       The  "wizard"  options are intended for experimentation with JPEG.  If you don't know what
       you are doing, don't use  them.   These  switches  are  documented  further  in  the  file
       wizard.doc that comes with the Independent JPEG Group's JPEG library.


       This  example  compresses  the  PPM file foo.ppm with a quality factor of 60 and saves the
       output as foo.jpg:

              pnmtojpeg --quality=60 foo.ppm > foo.jpg

              cat foo.bmp | bmptoppm | pnmtojpeg > foo.jpg


       JFIF is not ideal for cartoons, line drawings, and other  images  that  have  only  a  few
       distinct  colors.   For  those,  try instead pnmtopng or ppmtobmp.  If you need to convert
       such an image to JFIF, though,  you  should  experiment  with  pnmtojpeg's  --quality  and
       --smooth options to get a satisfactory conversion.  --smooth 10 or so is often helpful.

       JPEG  compression  is  notable  for  being  a  "lossy."  This means that, unlike with most
       graphics conversions, you lose information, which means image quality, when you convert to
       JFIF.   If  you  convert  from  PPM  to  JFIF and back repeatedly, image quality loss will
       accumulate.  After ten or so cycles the image may be noticeably worse than  it  was  after
       one cycle.

       Because  of  this,  you should do all the manipulation you have to do on the image in some
       other format and convert to JFIF as the last step.  And if you can  keep  a  copy  in  the
       original  format, so much the better.  PNG is a good choice for a format that is lossless,
       yet fairly compact.  GIF is another way to go, but chances are  you  can't  create  a  GIF
       image  without  owing  a  lot  of  money  to Unisys and IBM, holders of patents on the LZW
       compression used in the GIF format.

       The --optimize option to pnmtojpeg is 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 small JFIF files; the percentage improvement is often a lot more than it is on larger
       files.   (At  present,  --optimize  mode  is  automatically  in effect when you generate a
       progressive JPEG file).

       Another program, cjpeg, is similar.  cjpeg is maintained by the Independent JPEG Group and
       packaged  with  the  JPEG  library which pnmtojpeg uses for all its JPEG work.  Because of
       that, you may expect it to exploit more current JPEG features.  Also, since  you  have  to
       have  the  library  to  run  pnmtojpeg,  but  not  vice  versa, cjpeg may be more commonly

       On the other hand, cjpeg does not use the NetPBM libraries to process its  input,  as  all
       the NetPBM tools such as pnmtojpeg do.  This means it is less likely to be consistent with
       all the other programs that deal with the NetPBM formats.  Also,  the  command  syntax  of
       pnmtojpeg is consistent with that of the other Netpbm tools, unlike cjpeg.


       Use  the -scan option to specify a scan script.  Or use the -progressive option to specify
       a particular built-in scan script.

       Just what a scan script is, and the basic format of the scan script file,  is  covered  in
       the  wizard.doc  file  that  comes  with  the Independent JPEG Group's JPEG library.  Scan
       scripts are same for pnmtojpeg as the are for cjpeg.

       This section contains additional information that isn't, but probably should be,  in  that

       First,  there are many restrictions on what is a valid scan script.  The JPEG library, and
       thus pnmtojpeg, checks thoroughly for any lack of compliance with these restrictions,  but
       does little to tell you how the script fails to comply.  The messages are very general and
       sometimes untrue.

       To start with, the entries for the DC coefficient must come before any entries for the  AC
       coefficients.   The  DC  coefficient  is  Coefficient 0; all the other coefficients are AC
       coefficients.  So in an entry for the DC coefficient, the two numbers after the colon must
       be 0 and 0.  In an entry for AC coefficients, the first number after the colon must not be

       In a DC entry, the color components must be in increasing order.  E.g. "0,2,1" before  the
       colon is wrong.  So is "0,0,0".

       In  an  entry  for an AC coeffient, you must specify only one color component.  I.e. there
       can be only one number before the colon.

       In the first entry for a particular coefficient for a particular color component, the "Ah"
       value  must be zero, but the Al value can be any valid bit number.  In subsequent entries,
       Ah must be the Al value from the previous entry  (for  that  coefficient  for  that  color
       component), and the Al value must be one less than the Ah value.

       The  script  must  ultimately  specify  at least some of the DC coefficent for every color
       component.  Otherwise, you get the error message "Script does not transmit all the  data."
       You need not specify all of the bits of the DC coefficient, or any of the AC coefficients.

       There  is  a  standard option in building the JPEG library to omit scan script capability.
       If for some reason your library was built with this option, you get the message "Requested
       feature was omitted at compile time."


              If  this  environment  variable is set, its value is the default memory limit.  The
              value  is  specified  as  described  for  the  --maxmemory  option.   An   explicit
              --maxmemory option overrides any JPEGMEM.


       cjpeg(1), djpeg(1), jpegtran(1), rdjpgcom(1), wrjpgcom(1)
       ppm(5), pgm(5), jpegtopnm(1)
       Wallace,  Gregory K.  "The JPEG Still Picture Compression Standard", Communications of the
       ACM, April 1991 (vol. 34, no. 4), pp. 30-44.


       Arithmetic coding is not supported for legal reasons.

       The program could be much faster.


       pnmtojpeg and this man page were derived in large part from cjpeg, by the Independent JPEG
       Group.  The program is otherwise by Bryan Henderson on March 07, 2000.

                                          07 March 2000                              PNMTOJPEG(1)