<!DOCTYPE REFENTRY PUBLIC "-//Sun Microsystems//DTD DocBook V3.0-Based SolBook Subset V2.0//EN" [

<!--ArborText, Inc., 1988-1999, v.4002-->

<!ENTITY cmd "cjpeg">

<!ENTITY % commonents SYSTEM "smancommon.ent">

%commonents;

<!ENTITY % booktitles SYSTEM "booktitles.ent">

%booktitles;

<!ENTITY suncopy "Copyright (c) 2004, Sun Microsystems, Inc. All Rights Reserved.">

]>

<?Pub UDT _bookmark _target>

<?Pub Inc>

<refentry id="cjpeg-1">

<refmeta><refentrytitle>cjpeg</refentrytitle><manvolnum>1</manvolnum>

<refmiscinfo class="date">26 Mar 2004</refmiscinfo>

<refmiscinfo class="sectdesc">&man1;</refmiscinfo>

<refmiscinfo class="software">&release;</refmiscinfo>

<refmiscinfo class="arch">generic</refmiscinfo>

<refmiscinfo class="copyright">&suncopy;</refmiscinfo>

</refmeta>

<indexterm><primary>cjpeg</primary></indexterm><indexterm><primary>compress

an image file to a JPEG file</primary></indexterm>

<refnamediv id="cjpeg-1-name"><refname>cjpeg</refname><refpurpose>compress

an image file to a JPEG file</refpurpose></refnamediv>

<refsynopsisdiv id="cjpeg-1-synp"><title>&synp-tt;</title>

<cmdsynopsis><command>&cmd;</command><arg choice="opt"><option role="nodash"><replaceable>

options</replaceable></option></arg><arg choice="opt"><option role="nodash"><replaceable>

filename</replaceable></option></arg>

</cmdsynopsis></refsynopsisdiv>

<refsect1 id="cjpeg-1-desc"><title>&desc-tt;</title>

<para><command>&cmd;</command> compresses the named image file, or the standard

input if no file is named, and produces a JPEG/JFIF file on the standard output.

The following input file formats are currently supported:</para>

<itemizedlist>

<listitem><para>BMP</para></listitem>

<listitem><para>PGM, the PBMPLUS gray-scale format</para></listitem>

<listitem><para>PPM, the PBMPLUS color format</para></listitem>

<listitem><para>RLE, the Utah Raster Toolkit format</para></listitem>

<listitem><para>Targa</para></listitem>

</itemizedlist>

<para>RLE is supported only if the URT library is available.</para>

</refsect1>

<refsect1 id="cjpeg-1-opts"><title>&opts-tt;</title>

<para>All options may be abbreviated. For example, <option>grayscale</option>

may be written <option>gray</option> or <option>gr</option>. Upper and lower

case are equivalent. For example, <option>BMP</option> is the same as <option>

bmp</option>. British spellings are also accepted. For example, <option>greyscale

</option>.</para>

<refsect2 id="cjpeg-1-opts-basic">

<title>Basic Options</title>

<para>The following basic options are supported:</para>

<variablelist termlength="medium">

<varlistentry><term><option>grayscale</option></term><listitem><para>Create

a monochrome JPEG file from color input. You should use this switch when compressing

a grayscale BMP file, because <command>&cmd;</command> cannot detect whether

a BMP file uses only shades of gray. By specifying the <option>grayscale</option>

option, you create a smaller JPEG file that takes less time to process.</para>

</listitem></varlistentry>

<varlistentry><term><option>optimize</option></term><listitem><para>Optimize

the entropy encoding parameters. If you do not specify this option, default

encoding parameters are used. <option>optimize</option> usually makes the

JPEG file a little smaller, but <command>&cmd;</command> runs more slowly

and needs much more memory. Image quality and speed of decompression are unaffected

by the <option>optimize</option> option.</para>

</listitem></varlistentry>

<varlistentry><term><option>progressive</option></term><listitem><para>Create

a progressive JPEG file. For more information about the <option>progressive

</option> option, see the Extended Description section.</para>

</listitem></varlistentry>

<varlistentry><term><option>quality <replaceable>N</replaceable></option></term>

<listitem><para>Scale the quantization tables to adjust image quality. <replaceable>

N</replaceable> ranges from 0 (worst) to 100 (best). The default value is

75. For more information about the <option>quality</option> option, see the

Extended Description section.</para>

</listitem></varlistentry>

<varlistentry><term><option>targa</option></term><listitem><para>Specify that

the input file is in Targa format. Targa files that contain an "identification"

field are not automatically recognized by <command>&cmd;</command>. For such

files, you must specify <option>targa</option> to make <command>&cmd;</command>

treat the input as Targa format. For most Targa files, you do not need this

switch.</para>

</listitem></varlistentry>

</variablelist></refsect2>

<refsect2 id="cjpeg-1-opts-int">

<title>Intermediate Options</title>

<para>The following intermediate options are supported:</para>

<variablelist termlength="medium">

<varlistentry><term><option>dct fast</option></term><listitem><para>Use the

fast integer DCT method. This method is less accurate than the integer DCT

method or the floating-point DCT method.</para>

</listitem></varlistentry>

<varlistentry><term><option>dct float</option></term><listitem><para>Use the

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. The results of the floating-point method may vary

slightly across machines, while the integer methods should give the same results

everywhere.</para>

</listitem></varlistentry>

<varlistentry><term><option>dct int</option></term><listitem><para>Use the

integer DCT method. This is the default method.</para>

</listitem></varlistentry>

<varlistentry><term><option>maxmemory <replaceable>N</replaceable></option></term>

<listitem><para>Set the limit for the amount of memory to use in processing

large images. <replaceable>N</replaceable> is specified in thousands of bytes,

or in millions of bytes if "M" is specified with the number. For example, <option>

max 4m</option> selects 4000000 bytes. If more space is needed, temporary

files are used.</para>

</listitem></varlistentry>

<varlistentry><term><option>outfile <replaceable>name</replaceable></option></term>

<listitem><para>Send the output image to the named file, instead of to the

standard output.</para>

</listitem></varlistentry>

<varlistentry><term><option>restart <replaceable>N</replaceable></option></term>

<listitem><para>Emit a JPEG restart marker every <replaceable>N</replaceable>

MCU rows, or every <replaceable>N</replaceable> MCU blocks if "B" is specified

with the number. The default value is <option>restart 0</option>, which means

no restart markers. For more information about the <option>restart</option>

option, see the Extended Description section.</para>

</listitem></varlistentry>

<varlistentry><term><option>smooth <replaceable>N</replaceable></option></term>

<listitem><para>Smooth the input image to eliminate dithering noise. <replaceable>

N</replaceable> ranges from 1 to 100, and indicates the strength of smoothing.

The default value is <option>smooth 0</option>, which means no smoothing.

For more information about the <option>smooth</option> option, see the Extended

Description section.</para>

</listitem></varlistentry>

<varlistentry><term><option>verbose</option></term><listitem><para>Display

version information at startup, and enable debug printout. The <option>vv

</option> option displays more verbose output than the <option>v</option>

option. The <option>vvv</option> option displays the most verbose output.

</para>

<para>You can also use <option>debug</option> to specify this option.</para>

</listitem></varlistentry>

</variablelist></refsect2>

<refsect2 id="cjpeg-1-opts-adv">

<title>Advanced Options</title>

<para>The following advanced options are supported for advanced users only:

</para>

<variablelist termlength="medium">

<varlistentry><term><option>baseline</option></term><listitem><para>Force

baseline-compatible quantization tables to be generated. This clamps the quantization

values to 8 bits, even at low quality settings. This option is poorly named,

because <option>baseline</option> does not ensure that the output is actually

baseline JPEG. For example, you can use the <option>baseline</option> and <option>

progressive</option> options together.</para>

</listitem></varlistentry>

<varlistentry><term><option>qslots <replaceable>N</replaceable>[,...]</option></term>

<listitem><para>Select which quantization table to use for each color component.

</para>

</listitem></varlistentry>

<varlistentry><term><option>qtables <replaceable>file</replaceable></option></term>

<listitem><para>Use the quantization tables provided in the specified text

file.</para>

</listitem></varlistentry>

<varlistentry><term><option>sample HxV[,...]</option></term><listitem><para>

Set the JPEG sampling factors for each color component.</para>

</listitem></varlistentry>

<varlistentry><term><option>scans <replaceable>file</replaceable></option></term>

<listitem><para>Use the scan script provided in the specified text file.</para>

</listitem></varlistentry>

</variablelist></refsect2>

</refsect1>

<refsect1 id="cjpeg-1-oper"><title>&oper-tt;</title>

<para>The following operands are supported:</para>

<variablelist termlength="medium">

<varlistentry><term><option role="nodash"><replaceable>filename</replaceable></option></term>

<listitem><para>The name of the image file to be compressed.</para>

</listitem></varlistentry>

</variablelist></refsect1>

<refsect1 id="cjpeg-1-exde"><title>&exde-tt;</title>

<refsect2 id="cjpeg-1-exde-quality">

<title>The <option>quality</option> Option</title>

<para>The <option>quality</option> option enables you to trade compressed

file size against the quality of the reconstructed image: the higher the quality

setting, the larger the JPEG file, and the greater the similarity between

the output image and the original input. Normally, you use the lowest quality

setting that decompresses into an output image that is visually indistinguishable

from the original image. For this purpose, the quality setting should be between

50 and 95. The default value of 75 is often just right. If you see defects

at <option>quality 75</option>, increase the quality by 5 until you are happy

with the output image. The optimal setting varies from one image to another.

</para>

<para>A value of <option>quality 100</option> generates a quantization table

of ones. This minimizes loss in the quantization step, but information is

still lost in subsampling, as well as roundoff error. The <option>quality

100</option> setting is mainly of interest for experimental purposes. Quality

values above 95 are not recommended for normal use, as the compressed file

size increases dramatically for very little gain in output image quality.

</para>

<para>Quality values below 50 produce very small files of low image quality.

Settings of 5 to 10 might be useful in preparing an index of a large image

library, for example. Quality values below 25 generate 2-byte quantization

tables, which are considered optional in the JPEG standard. <command>&cmd;</command>

emits a warning message when you specify such a quality value, because some

other JPEG programs might be unable to decode the resulting file. Use <option>

baseline</option> if you need to ensure compatibility at low quality values.

</para>

</refsect2>

<refsect2 id="cjpeg-1-exde-prog">

<title>The <option>progressive</option> Option</title>

<para>The <option>progressive</option> option creates a "progressive JPEG"

file. In this type of JPEG file, the data is stored in multiple scans of increasing

quality. If the file is 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 JPEG file of the same quality setting, and the total

file size is about the same, or a little smaller. Caution: progressive JPEG

is not yet widely implemented, so many decoders are unable to view a progressive

JPEG file at all.</para>

</refsect2>

<refsect2 id="cjpeg-1-exde-restart">

<title>The <option>restart</option> Option</title>

<para>The <option>restart</option> option inserts extra markers that allow

a JPEG decoder to resynchronize after a transmission error. Without restart

markers, any damage to a compressed file usually ruins 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 from the point of the error

to the next restart marker. The restart markers occupy extra space. We recommend <option>

restart 1</option> for images that are transmitted across unreliable networks.

</para>

</refsect2>

<refsect2 id="cjpeg-1-exde-smooth">

<title>The <option>smooth</option> Option</title>

<para>The <option>smooth</option> option filters the input to eliminate fine-scale

noise. This option is often useful when you convert dithered images to JPEG:

a moderate smoothing factor of 10 to 50 deletes dithering patterns from the

input file, resulting in a smaller JPEG file and a better-looking image. If

the smoothing factor is too large, the image visibly blurs.</para>

</refsect2>

<refsect2 id="cjpeg-1-exde-hints">

<title>Hints</title>

<para>Color GIF files are not the ideal input for JPEG. JPEG is really intended

for the compression of full-color 24-bit images. In particular, do not try

to convert cartoons, line drawings, or other images that have only a few distinct

colors. GIF works well on these, but JPEG does not. If you want to convert

a GIF file to JPEG, you should experiment with the <option>quality</option>

and <option>smooth</option> options to get a satisfactory conversion. A value

of <option>smooth 10</option> is often helpful.</para>

<para>Avoid running an image through a series of JPEG compression/decompression

cycles. Image quality loss will accumulate. After ten cycles, the image may

be noticeably worse than it was after one cycle. Use a lossless format while

manipulating an image, then convert to JPEG format when you are ready to file

the image away. </para>

<para>Use the <option>optimize</option> option when you make a "final" version

for posting or archiving. Also, when you use low quality settings, make very

small JPEG files. The percentage improvement is often much greater than on

larger files. At present, the <option>optimize</option> option is always selected

when generating progressive JPEG files.</para>

</refsect2>

</refsect1>

<refsect1 id="cjpeg-1-exam"><title>&exam-tt;</title>

<example role="example">

<title>Compressing the PPM File test.ppm With a Quality Factor of 60 and Saving

the Output as test.jpg</title>

<para><screen>example% <userinput>&cmd; -quality 60 test.ppm > test.jpg

</userinput></screen></para>

</example>

</refsect1>

<refsect1 id="cjpeg-1-envr"><title>&envr-tt;</title>

<para><command>&cmd;</command> uses the following environment variables:</para>

<variablelist>

<varlistentry><term>JPEGMEM</term><listitem><para>The value of this environment

variable, if set, is the default memory limit. The value is specified as described

for the <option>maxmemory</option> option. JPEGMEM overrides the default value

specified when the program was compiled, and is in turn overridden by an explicit <option>

maxmemory</option> option.</para>

</listitem></varlistentry>

</variablelist></refsect1>

<refsect1 id="cjpeg-1-attr"><title>&attr-tt;</title>

<para>See <olink targetdocent="REFMAN5" localinfo="attributes-5"><citerefentry>

<refentrytitle>attributes</refentrytitle><manvolnum>5</manvolnum></citerefentry></olink>

for descriptions of the following attributes:</para>

<informaltable frame="all">

<tgroup cols="2" colsep="1" rowsep="1"><colspec colname="COLSPEC0" colwidth="1*">

<colspec colname="COLSPEC1" colwidth="1*">

<thead>

<row><entry align="center" valign="middle">ATTRIBUTE TYPE</entry><entry align="center"

valign="middle">ATTRIBUTE VALUE</entry></row>

</thead>

<tbody>

<row><entry><para>Availability</para></entry><entry><para>image/library/libjpeg</para></entry>

</row>

<row><entry colname="COLSPEC0"><para>Interface stability</para></entry><entry

colname="COLSPEC1"><para>Uncommitted</para></entry></row>

</tbody>

</tgroup>

</informaltable>

</refsect1>

<refsect1 id="cjpeg-1-also"><title>&also-tt;</title>

<para>Wallace, Gregory K., <citetitle>The JPEG Still Picture Compression Standard

</citetitle> Communications of the ACM, April 1991 (vol. 34, no. 4), pp. 30-44.

</para>

<para><citerefentry><refentrytitle>djpeg</refentrytitle><manvolnum>1</manvolnum>

</citerefentry>, <citerefentry><refentrytitle>jpegtran</refentrytitle><manvolnum>

1</manvolnum></citerefentry>, <citerefentry><refentrytitle>rdjpgcom</refentrytitle>

<manvolnum>1</manvolnum></citerefentry>, <citerefentry><refentrytitle>wrjpgcom

</refentrytitle><manvolnum>1</manvolnum></citerefentry></para>

</refsect1>

<refsect1 id="cjpeg-1-note"><title>&note-tt;</title>

<para>Arithmetic coding is not supported. GIF input files are not supported.

Not all variants of BMP and Targa file formats are supported.</para>

<para>This man page was originally written by the Independent JPEG Group.

Updated by Breda McColgan, Sun Microsystems Inc., 2004. </para>

</refsect1>

</refentry>