<!DOCTYPE html>
<!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
<!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
<head>
<meta charset="utf-8">
<meta name="viewport" content="width=device-width, initial-scale=1.0">
<title>2.7.0 — Pillow (PIL Fork) 5.4.1 documentation</title>
<script type="text/javascript" src="../_static/js/modernizr.min.js"></script>
<script type="text/javascript" id="documentation_options" data-url_root="../" src="../_static/documentation_options.js"></script>
<script type="text/javascript" src="../_static/jquery.js"></script>
<script type="text/javascript" src="../_static/underscore.js"></script>
<script type="text/javascript" src="../_static/doctools.js"></script>
<script type="text/javascript" src="../_static/language_data.js"></script>
<script type="text/javascript" src="../_static/js/script.js"></script>
<script type="text/javascript" src="../_static/js/theme.js"></script>
<link rel="stylesheet" href="../_static/css/theme.css" type="text/css" />
<link rel="stylesheet" href="../_static/pygments.css" type="text/css" />
<link rel="author" title="About these documents" href="../about.html" />
<link rel="index" title="Index" href="../genindex.html" />
<link rel="search" title="Search" href="../search.html" />
<link rel="next" title="Deprecations and removals" href="../deprecations.html" />
<link rel="prev" title="2.8.0" href="2.8.0.html" />
</head>
<body class="wy-body-for-nav">
<div class="wy-grid-for-nav">
<nav data-toggle="wy-nav-shift" class="wy-nav-side">
<div class="wy-side-scroll">
<div class="wy-side-nav-search" >
<a href="../index.html" class="icon icon-home"> Pillow (PIL Fork)
</a>
<div class="version">
5.4.1
</div>
<div role="search">
<form id="rtd-search-form" class="wy-form" action="../search.html" method="get">
<input type="text" name="q" placeholder="Search docs" />
<input type="hidden" name="check_keywords" value="yes" />
<input type="hidden" name="area" value="default" />
</form>
</div>
</div>
<div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
<ul class="current">
<li class="toctree-l1"><a class="reference internal" href="../installation.html">Installation</a></li>
<li class="toctree-l1"><a class="reference internal" href="../handbook/index.html">Handbook</a></li>
<li class="toctree-l1"><a class="reference internal" href="../reference/index.html">Reference</a></li>
<li class="toctree-l1"><a class="reference internal" href="../porting.html">Porting</a></li>
<li class="toctree-l1"><a class="reference internal" href="../about.html">About</a></li>
<li class="toctree-l1 current"><a class="reference internal" href="index.html">Release Notes</a><ul class="current">
<li class="toctree-l2"><a class="reference internal" href="5.4.1.html">5.4.1</a></li>
<li class="toctree-l2"><a class="reference internal" href="5.4.0.html">5.4.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="5.3.0.html">5.3.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="5.2.0.html">5.2.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="5.1.0.html">5.1.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="5.0.0.html">5.0.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="4.3.0.html">4.3.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="4.2.1.html">4.2.1</a></li>
<li class="toctree-l2"><a class="reference internal" href="4.2.0.html">4.2.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="4.1.1.html">4.1.1</a></li>
<li class="toctree-l2"><a class="reference internal" href="4.1.0.html">4.1.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="4.0.0.html">4.0.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="3.4.0.html">3.4.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="3.3.2.html">3.3.2</a></li>
<li class="toctree-l2"><a class="reference internal" href="3.3.0.html">3.3.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="3.2.0.html">3.2.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="3.1.2.html">3.1.2</a></li>
<li class="toctree-l2"><a class="reference internal" href="3.1.1.html">3.1.1</a></li>
<li class="toctree-l2"><a class="reference internal" href="3.1.0.html">3.1.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="3.0.0.html">3.0.0</a></li>
<li class="toctree-l2"><a class="reference internal" href="2.8.0.html">2.8.0</a></li>
<li class="toctree-l2 current"><a class="current reference internal" href="#">2.7.0</a><ul>
<li class="toctree-l3"><a class="reference internal" href="#sane-plugin">Sane Plugin</a></li>
<li class="toctree-l3"><a class="reference internal" href="#png-text-chunk-size-limits">Png text chunk size limits</a></li>
<li class="toctree-l3"><a class="reference internal" href="#image-resizing-filters">Image resizing filters</a><ul>
<li class="toctree-l4"><a class="reference internal" href="#bicubic-and-bilinear-downscaling">Bicubic and bilinear downscaling</a></li>
<li class="toctree-l4"><a class="reference internal" href="#antialias-renamed-to-lanczos">Antialias renamed to Lanczos</a></li>
<li class="toctree-l4"><a class="reference internal" href="#lanczos-upscaling-quality">Lanczos upscaling quality</a></li>
<li class="toctree-l4"><a class="reference internal" href="#bicubic-upscaling-quality">Bicubic upscaling quality</a></li>
<li class="toctree-l4"><a class="reference internal" href="#resize-performance">Resize performance</a></li>
<li class="toctree-l4"><a class="reference internal" href="#default-filter-for-thumbnails">Default filter for thumbnails</a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#image-transposition">Image transposition</a></li>
<li class="toctree-l3"><a class="reference internal" href="#gaussian-blur-and-unsharp-mask">Gaussian blur and unsharp mask</a><ul>
<li class="toctree-l4"><a class="reference internal" href="#blur-radius">Blur radius</a></li>
<li class="toctree-l4"><a class="reference internal" href="#blur-performance">Blur performance</a></li>
<li class="toctree-l4"><a class="reference internal" href="#blur-quality">Blur quality</a></li>
</ul>
</li>
<li class="toctree-l3"><a class="reference internal" href="#tiff-parameter-changes">TIFF Parameter Changes</a></li>
</ul>
</li>
</ul>
</li>
<li class="toctree-l1"><a class="reference internal" href="../deprecations.html">Deprecations and removals</a></li>
</ul>
</div>
</div>
</nav>
<section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
<nav class="wy-nav-top" aria-label="top navigation">
<i data-toggle="wy-nav-top" class="fa fa-bars"></i>
<a href="../index.html">Pillow (PIL Fork)</a>
</nav>
<div class="wy-nav-content">
<div class="rst-content">
<div role="navigation" aria-label="breadcrumbs navigation">
<ul class="wy-breadcrumbs">
<li><a href="../index.html">Docs</a> »</li>
<li><a href="index.html">Release Notes</a> »</li>
<li>2.7.0</li>
<li class="wy-breadcrumbs-aside">
<a href="../_sources/releasenotes/2.7.0.rst.txt" rel="nofollow"> View page source</a>
</li>
</ul>
<hr/>
</div>
<div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
<div itemprop="articleBody">
<div class="section" id="id1">
<h1>2.7.0<a class="headerlink" href="#id1" title="Permalink to this headline">¶</a></h1>
<div class="section" id="sane-plugin">
<h2>Sane Plugin<a class="headerlink" href="#sane-plugin" title="Permalink to this headline">¶</a></h2>
<p>The Sane plugin has now been split into its own repo:
<a class="reference external" href="https://github.com/python-pillow/Sane">https://github.com/python-pillow/Sane</a> .</p>
</div>
<div class="section" id="png-text-chunk-size-limits">
<h2>Png text chunk size limits<a class="headerlink" href="#png-text-chunk-size-limits" title="Permalink to this headline">¶</a></h2>
<p>To prevent potential denial of service attacks using compressed text
chunks, there are now limits to the decompressed size of text chunks
decoded from PNG images. If the limits are exceeded when opening a PNG
image a <code class="docutils literal notranslate"><span class="pre">ValueError</span></code> will be raised.</p>
<p>Individual text chunks are limited to
<code class="xref py py-attr docutils literal notranslate"><span class="pre">PIL.PngImagePlugin.MAX_TEXT_CHUNK</span></code>, set to 1MB by
default. The total decompressed size of all text chunks is limited to
<code class="xref py py-attr docutils literal notranslate"><span class="pre">PIL.PngImagePlugin.MAX_TEXT_MEMORY</span></code>, which defaults to
64MB. These values can be changed prior to opening PNG images if you
know that there are large text blocks that are desired.</p>
</div>
<div class="section" id="image-resizing-filters">
<h2>Image resizing filters<a class="headerlink" href="#image-resizing-filters" title="Permalink to this headline">¶</a></h2>
<p>Image resizing methods <a class="reference internal" href="../reference/Image.html#PIL.Image.Image.resize" title="PIL.Image.Image.resize"><code class="xref py py-meth docutils literal notranslate"><span class="pre">resize()</span></code></a> and
<a class="reference internal" href="../reference/Image.html#PIL.Image.Image.thumbnail" title="PIL.Image.Image.thumbnail"><code class="xref py py-meth docutils literal notranslate"><span class="pre">thumbnail()</span></code></a> take a <cite>resample</cite> argument, which tells
which filter should be used for resampling. Possible values are:
<code class="xref py py-attr docutils literal notranslate"><span class="pre">PIL.Image.NEAREST</span></code>, <code class="xref py py-attr docutils literal notranslate"><span class="pre">PIL.Image.BILINEAR</span></code>,
<code class="xref py py-attr docutils literal notranslate"><span class="pre">PIL.Image.BICUBIC</span></code> and <code class="xref py py-attr docutils literal notranslate"><span class="pre">PIL.Image.ANTIALIAS</span></code>.
Almost all of them were changed in this version.</p>
<div class="section" id="bicubic-and-bilinear-downscaling">
<h3>Bicubic and bilinear downscaling<a class="headerlink" href="#bicubic-and-bilinear-downscaling" title="Permalink to this headline">¶</a></h3>
<p>From the beginning <code class="xref py py-attr docutils literal notranslate"><span class="pre">BILINEAR</span></code> and
<code class="xref py py-attr docutils literal notranslate"><span class="pre">BICUBIC</span></code> filters were based on affine transformations
and used a fixed number of pixels from the source image for every destination
pixel (2x2 pixels for <code class="xref py py-attr docutils literal notranslate"><span class="pre">BILINEAR</span></code> and 4x4 for
<code class="xref py py-attr docutils literal notranslate"><span class="pre">BICUBIC</span></code>). This gave an unsatisfactory result for
downscaling. At the same time, a high quality convolutions-based algorithm with
flexible kernel was used for <code class="xref py py-attr docutils literal notranslate"><span class="pre">ANTIALIAS</span></code> filter.</p>
<p>Starting from Pillow 2.7.0, a high quality convolutions-based algorithm is used
for all of these three filters.</p>
<p>If you have previously used any tricks to maintain quality when downscaling with
<code class="xref py py-attr docutils literal notranslate"><span class="pre">BILINEAR</span></code> and <code class="xref py py-attr docutils literal notranslate"><span class="pre">BICUBIC</span></code> filters
(for example, reducing within several steps), they are unnecessary now.</p>
</div>
<div class="section" id="antialias-renamed-to-lanczos">
<h3>Antialias renamed to Lanczos<a class="headerlink" href="#antialias-renamed-to-lanczos" title="Permalink to this headline">¶</a></h3>
<p>A new <code class="xref py py-attr docutils literal notranslate"><span class="pre">PIL.Image.LANCZOS</span></code> constant was added instead of
<code class="xref py py-attr docutils literal notranslate"><span class="pre">ANTIALIAS</span></code>.</p>
<p>When <code class="xref py py-attr docutils literal notranslate"><span class="pre">ANTIALIAS</span></code> was initially added, it was the only
high-quality filter based on convolutions. It’s name was supposed to reflect
this. Starting from Pillow 2.7.0 all resize method are based on convolutions.
All of them are antialias from now on. And the real name of the
<code class="xref py py-attr docutils literal notranslate"><span class="pre">ANTIALIAS</span></code> filter is Lanczos filter.</p>
<p>The <code class="xref py py-attr docutils literal notranslate"><span class="pre">ANTIALIAS</span></code> constant is left for backward compatibility
and is an alias for <code class="xref py py-attr docutils literal notranslate"><span class="pre">LANCZOS</span></code>.</p>
</div>
<div class="section" id="lanczos-upscaling-quality">
<h3>Lanczos upscaling quality<a class="headerlink" href="#lanczos-upscaling-quality" title="Permalink to this headline">¶</a></h3>
<p>The image upscaling quality with <code class="xref py py-attr docutils literal notranslate"><span class="pre">LANCZOS</span></code> filter was
almost the same as <code class="xref py py-attr docutils literal notranslate"><span class="pre">BILINEAR</span></code> due to bug. This has been fixed.</p>
</div>
<div class="section" id="bicubic-upscaling-quality">
<h3>Bicubic upscaling quality<a class="headerlink" href="#bicubic-upscaling-quality" title="Permalink to this headline">¶</a></h3>
<p>The <code class="xref py py-attr docutils literal notranslate"><span class="pre">BICUBIC</span></code> filter for affine transformations produced
sharp, slightly pixelated image for upscaling. Bicubic for convolutions is
more soft.</p>
</div>
<div class="section" id="resize-performance">
<h3>Resize performance<a class="headerlink" href="#resize-performance" title="Permalink to this headline">¶</a></h3>
<p>In most cases, convolution is more a expensive algorithm for downscaling
because it takes into account all the pixels of source image. Therefore
<code class="xref py py-attr docutils literal notranslate"><span class="pre">BILINEAR</span></code> and <code class="xref py py-attr docutils literal notranslate"><span class="pre">BICUBIC</span></code> filters’
performance can be lower than before. On the other hand the quality of
<code class="xref py py-attr docutils literal notranslate"><span class="pre">BILINEAR</span></code> and <code class="xref py py-attr docutils literal notranslate"><span class="pre">BICUBIC</span></code> was close to
<code class="xref py py-attr docutils literal notranslate"><span class="pre">NEAREST</span></code>. So if such quality is suitable for your tasks
you can switch to <code class="xref py py-attr docutils literal notranslate"><span class="pre">NEAREST</span></code> filter for downscaling,
which will give a huge improvement in performance.</p>
<p>At the same time performance of convolution resampling for downscaling has been
improved by around a factor of two compared to the previous version.
The upscaling performance of the <code class="xref py py-attr docutils literal notranslate"><span class="pre">LANCZOS</span></code> filter has
remained the same. For <code class="xref py py-attr docutils literal notranslate"><span class="pre">BILINEAR</span></code> filter it has improved by
1.5 times and for <code class="xref py py-attr docutils literal notranslate"><span class="pre">BICUBIC</span></code> by four times.</p>
</div>
<div class="section" id="default-filter-for-thumbnails">
<h3>Default filter for thumbnails<a class="headerlink" href="#default-filter-for-thumbnails" title="Permalink to this headline">¶</a></h3>
<p>In Pillow 2.5 the default filter for <a class="reference internal" href="../reference/Image.html#PIL.Image.Image.thumbnail" title="PIL.Image.Image.thumbnail"><code class="xref py py-meth docutils literal notranslate"><span class="pre">thumbnail()</span></code></a> was
changed from <code class="xref py py-attr docutils literal notranslate"><span class="pre">NEAREST</span></code> to <code class="xref py py-attr docutils literal notranslate"><span class="pre">ANTIALIAS</span></code>.
Antialias was chosen because all the other filters gave poor quality for
reduction. Starting from Pillow 2.7.0, <code class="xref py py-attr docutils literal notranslate"><span class="pre">ANTIALIAS</span></code> has been
replaced with <code class="xref py py-attr docutils literal notranslate"><span class="pre">BICUBIC</span></code>, because it’s faster and
<code class="xref py py-attr docutils literal notranslate"><span class="pre">ANTIALIAS</span></code> doesn’t give any advantages after
downscaling with libjpeg, which uses supersampling internally, not convolutions.</p>
</div>
</div>
<div class="section" id="image-transposition">
<h2>Image transposition<a class="headerlink" href="#image-transposition" title="Permalink to this headline">¶</a></h2>
<p>A new method <code class="xref py py-attr docutils literal notranslate"><span class="pre">PIL.Image.TRANSPOSE</span></code> has been added for the
<a class="reference internal" href="../reference/Image.html#PIL.Image.Image.transpose" title="PIL.Image.Image.transpose"><code class="xref py py-meth docutils literal notranslate"><span class="pre">transpose()</span></code></a> operation in addition to
<code class="xref py py-attr docutils literal notranslate"><span class="pre">FLIP_LEFT_RIGHT</span></code>, <code class="xref py py-attr docutils literal notranslate"><span class="pre">FLIP_TOP_BOTTOM</span></code>,
<code class="xref py py-attr docutils literal notranslate"><span class="pre">ROTATE_90</span></code>, <code class="xref py py-attr docutils literal notranslate"><span class="pre">ROTATE_180</span></code>,
<code class="xref py py-attr docutils literal notranslate"><span class="pre">ROTATE_270</span></code>. <code class="xref py py-attr docutils literal notranslate"><span class="pre">TRANSPOSE</span></code> is an algebra
transpose, with an image reflected across its main diagonal.</p>
<p>The speed of <code class="xref py py-attr docutils literal notranslate"><span class="pre">ROTATE_90</span></code>, <code class="xref py py-attr docutils literal notranslate"><span class="pre">ROTATE_270</span></code>
and <code class="xref py py-attr docutils literal notranslate"><span class="pre">TRANSPOSE</span></code> has been significantly improved for large
images which don’t fit in the processor cache.</p>
</div>
<div class="section" id="gaussian-blur-and-unsharp-mask">
<h2>Gaussian blur and unsharp mask<a class="headerlink" href="#gaussian-blur-and-unsharp-mask" title="Permalink to this headline">¶</a></h2>
<p>The <a class="reference internal" href="../reference/ImageFilter.html#PIL.ImageFilter.GaussianBlur" title="PIL.ImageFilter.GaussianBlur"><code class="xref py py-meth docutils literal notranslate"><span class="pre">GaussianBlur()</span></code></a> implementation has been replaced
with a sequential application of box filters. The new implementation is based on
“Theoretical foundations of Gaussian convolution by extended box filtering” from
the Mathematical Image Analysis Group. As <a class="reference internal" href="../reference/ImageFilter.html#PIL.ImageFilter.UnsharpMask" title="PIL.ImageFilter.UnsharpMask"><code class="xref py py-meth docutils literal notranslate"><span class="pre">UnsharpMask()</span></code></a>
implementations use Gaussian blur internally, all changes from this chapter
are also applicable to it.</p>
<div class="section" id="blur-radius">
<h3>Blur radius<a class="headerlink" href="#blur-radius" title="Permalink to this headline">¶</a></h3>
<p>There was an error in the previous version of Pillow, where blur radius (the
standard deviation of Gaussian) actually meant blur diameter. For example, to
blur an image with actual radius 5 you were forced to use value 10. This has
been fixed. Now the meaning of the radius is the same as in other software.</p>
<p>If you used a Gaussian blur with some radius value, you need to divide this
value by two.</p>
</div>
<div class="section" id="blur-performance">
<h3>Blur performance<a class="headerlink" href="#blur-performance" title="Permalink to this headline">¶</a></h3>
<p>Box filter computation time is constant relative to the radius and depends
on source image size only. Because the new Gaussian blur implementation
is based on box filter, its computation time also doesn’t depend on the blur
radius.</p>
<p>For example, previously, if the execution time for a given test image was 1
second for radius 1, 3.6 seconds for radius 10 and 17 seconds for 50, now blur
with any radius on same image is executed for 0.2 seconds.</p>
</div>
<div class="section" id="blur-quality">
<h3>Blur quality<a class="headerlink" href="#blur-quality" title="Permalink to this headline">¶</a></h3>
<p>The previous implementation takes into account only source pixels within
2 * standard deviation radius for every destination pixel. This was not enough,
so the quality was worse compared to other Gaussian blur software.</p>
<p>The new implementation does not have this drawback.</p>
</div>
</div>
<div class="section" id="tiff-parameter-changes">
<h2>TIFF Parameter Changes<a class="headerlink" href="#tiff-parameter-changes" title="Permalink to this headline">¶</a></h2>
<p>Several kwarg parameters for saving TIFF images were previously
specified as strings with included spaces (e.g. ‘x resolution’). This
was difficult to use as kwargs without constructing and passing a
dictionary. These parameters now use the underscore character instead
of space. (e.g. ‘x_resolution’)</p>
</div>
</div>
</div>
</div>
<footer>
<div class="rst-footer-buttons" role="navigation" aria-label="footer navigation">
<a href="../deprecations.html" class="btn btn-neutral float-right" title="Deprecations and removals" accesskey="n" rel="next">Next <span class="fa fa-arrow-circle-right"></span></a>
<a href="2.8.0.html" class="btn btn-neutral float-left" title="2.8.0" accesskey="p" rel="prev"><span class="fa fa-arrow-circle-left"></span> Previous</a>
</div>
<hr/>
<div role="contentinfo">
<p>
© Copyright 1995-2011 Fredrik Lundh, 2010-2018 Alex Clark and Contributors
</p>
</div>
Built with <a href="http://sphinx-doc.org/">Sphinx</a> using a <a href="https://github.com/rtfd/sphinx_rtd_theme">theme</a> provided by <a href="https://readthedocs.org">Read the Docs</a>.
</footer>
</div>
</div>
</section>
</div>
<script type="text/javascript">
jQuery(function () {
SphinxRtdTheme.Navigation.enable(true);
});
</script>
</body>
</html>
