noise.html 44 KB

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960616263646566676869707172737475767778798081828384858687888990919293949596979899100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563564565566567568569570571572573574575576577578579580581582583584585586587588
  1. <!DOCTYPE html>
  2. <!--[if IE 8]><html class="no-js lt-ie9" lang="en" > <![endif]-->
  3. <!--[if gt IE 8]><!--> <html class="no-js" lang="en" > <!--<![endif]-->
  4. <head>
  5. <meta charset="utf-8">
  6. <meta name="viewport" content="width=device-width, initial-scale=1.0">
  7. <title>ipfml.filters.noise &mdash; ipfml v0.4.2 documentation</title>
  8. <link rel="stylesheet" href="../../../_static/css/theme.css" type="text/css" />
  9. <link rel="stylesheet" href="../../../_static/pygments.css" type="text/css" />
  10. <link rel="index" title="Index" href="../../../genindex.html" />
  11. <link rel="search" title="Search" href="../../../search.html" />
  12. <script src="../../../_static/js/modernizr.min.js"></script>
  13. </head>
  14. <body class="wy-body-for-nav">
  15. <div class="wy-grid-for-nav">
  16. <nav data-toggle="wy-nav-shift" class="wy-nav-side">
  17. <div class="wy-side-scroll">
  18. <div class="wy-side-nav-search">
  19. <a href="../../../index.html" class="icon icon-home"> ipfml
  20. </a>
  21. <div class="version">
  22. 0.4.2
  23. </div>
  24. <div role="search">
  25. <form id="rtd-search-form" class="wy-form" action="../../../search.html" method="get">
  26. <input type="text" name="q" placeholder="Search docs" />
  27. <input type="hidden" name="check_keywords" value="yes" />
  28. <input type="hidden" name="area" value="default" />
  29. </form>
  30. </div>
  31. </div>
  32. <div class="wy-menu wy-menu-vertical" data-spy="affix" role="navigation" aria-label="main navigation">
  33. <p class="caption"><span class="caption-text">Contents:</span></p>
  34. <ul>
  35. <li class="toctree-l1"><a class="reference internal" href="../../../description.html">Description</a></li>
  36. <li class="toctree-l1"><a class="reference internal" href="../../../ipfml.html">Documentation</a></li>
  37. <li class="toctree-l1"><a class="reference internal" href="../../../examples.html">Examples</a></li>
  38. <li class="toctree-l1"><a class="reference internal" href="../../../contributing.html">Contributing</a></li>
  39. </ul>
  40. </div>
  41. </div>
  42. </nav>
  43. <section data-toggle="wy-nav-shift" class="wy-nav-content-wrap">
  44. <nav class="wy-nav-top" aria-label="top navigation">
  45. <i data-toggle="wy-nav-top" class="fa fa-bars"></i>
  46. <a href="../../../index.html">ipfml</a>
  47. </nav>
  48. <div class="wy-nav-content">
  49. <div class="rst-content">
  50. <div role="navigation" aria-label="breadcrumbs navigation">
  51. <ul class="wy-breadcrumbs">
  52. <li><a href="../../../index.html">Docs</a> &raquo;</li>
  53. <li><a href="../../index.html">Module code</a> &raquo;</li>
  54. <li>ipfml.filters.noise</li>
  55. <li class="wy-breadcrumbs-aside">
  56. </li>
  57. </ul>
  58. <hr/>
  59. </div>
  60. <div role="main" class="document" itemscope="itemscope" itemtype="http://schema.org/Article">
  61. <div itemprop="articleBody">
  62. <h1>Source code for ipfml.filters.noise</h1><div class="highlight"><pre>
  63. <span></span><span class="sd">&quot;&quot;&quot;</span>
  64. <span class="sd">Noise filters to apply on images</span>
  65. <span class="sd">&quot;&quot;&quot;</span>
  66. <span class="c1"># main imports</span>
  67. <span class="kn">import</span> <span class="nn">numpy</span> <span class="k">as</span> <span class="nn">np</span>
  68. <span class="kn">import</span> <span class="nn">random</span>
  69. <span class="k">def</span> <span class="nf">_normalise</span><span class="p">(</span><span class="n">x</span><span class="p">):</span>
  70. <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">):</span>
  71. <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="n">_normalise</span><span class="p">,</span> <span class="n">x</span><span class="p">)))</span>
  72. <span class="k">if</span> <span class="n">x</span> <span class="o">&gt;</span> <span class="mi">255</span><span class="p">:</span>
  73. <span class="k">return</span> <span class="mi">255</span>
  74. <span class="k">if</span> <span class="n">x</span> <span class="o">&lt;</span> <span class="mi">0</span><span class="p">:</span>
  75. <span class="k">return</span> <span class="mi">0</span>
  76. <span class="k">return</span> <span class="n">x</span>
  77. <span class="k">def</span> <span class="nf">_global_noise_filter</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">generator</span><span class="p">,</span> <span class="n">updator</span><span class="p">,</span> <span class="n">identical</span><span class="o">=</span><span class="kc">False</span><span class="p">):</span>
  78. <span class="sd">&quot;&quot;&quot;White noise filter to apply on image</span>
  79. <span class="sd"> Args:</span>
  80. <span class="sd"> image: image used as input (2D or 3D image representation)</span>
  81. <span class="sd"> generator: lambda function used to generate random numpy array with specific distribution</span>
  82. <span class="sd"> updator: lambda function used to update pixel value</span>
  83. <span class="sd"> identical: keep or not identical noise distribution for each canal if RGB Image (default False)</span>
  84. <span class="sd"> Returns:</span>
  85. <span class="sd"> 2D Numpy array with specified noise applied</span>
  86. <span class="sd"> Example:</span>
  87. <span class="sd"> &gt;&gt;&gt; from ipfml.filters.noise import _global_noise_filter as gf</span>
  88. <span class="sd"> &gt;&gt;&gt; import numpy as np</span>
  89. <span class="sd"> &gt;&gt;&gt; image = np.random.uniform(0, 255, 10000).reshape((100, 100))</span>
  90. <span class="sd"> &gt;&gt;&gt; generator = lambda h, w: np.random.uniform(-0.5, 0.5, (h, w))</span>
  91. <span class="sd"> &gt;&gt;&gt; n = 10</span>
  92. <span class="sd"> &gt;&gt;&gt; k = 0.2</span>
  93. <span class="sd"> &gt;&gt;&gt; updator = lambda x, noise: x + n * k * noise</span>
  94. <span class="sd"> &gt;&gt;&gt; noisy_image = gf(image, generator, updator)</span>
  95. <span class="sd"> &gt;&gt;&gt; noisy_image.shape</span>
  96. <span class="sd"> (100, 100)</span>
  97. <span class="sd"> &quot;&quot;&quot;</span>
  98. <span class="n">image_array</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">asarray</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
  99. <span class="n">nb_chanel</span> <span class="o">=</span> <span class="mi">1</span>
  100. <span class="k">if</span> <span class="n">image_array</span><span class="o">.</span><span class="n">ndim</span> <span class="o">!=</span> <span class="mi">3</span><span class="p">:</span>
  101. <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">image_array</span><span class="o">.</span><span class="n">shape</span>
  102. <span class="k">else</span><span class="p">:</span>
  103. <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">nb_chanel</span> <span class="o">=</span> <span class="n">image_array</span><span class="o">.</span><span class="n">shape</span>
  104. <span class="k">if</span> <span class="n">nb_chanel</span> <span class="o">==</span> <span class="mi">1</span> <span class="ow">or</span> <span class="n">identical</span><span class="p">:</span>
  105. <span class="n">noise_filter</span> <span class="o">=</span> <span class="n">generator</span><span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
  106. <span class="c1"># final output numpy array</span>
  107. <span class="n">output_array</span> <span class="o">=</span> <span class="p">[]</span>
  108. <span class="c1"># check number of chanel</span>
  109. <span class="k">if</span> <span class="n">nb_chanel</span> <span class="o">==</span> <span class="mi">1</span><span class="p">:</span>
  110. <span class="n">noisy_image</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="n">updator</span><span class="p">,</span> <span class="n">image_array</span><span class="p">,</span> <span class="n">noise_filter</span><span class="p">)))</span>
  111. <span class="c1"># normalise values</span>
  112. <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="n">_normalise</span><span class="p">,</span> <span class="n">noisy_image</span><span class="p">)),</span> <span class="s1">&#39;uint8&#39;</span><span class="p">)</span>
  113. <span class="k">else</span><span class="p">:</span>
  114. <span class="c1"># final output numpy array</span>
  115. <span class="n">output_array</span> <span class="o">=</span> <span class="p">[]</span>
  116. <span class="k">for</span> <span class="n">chanel</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="n">nb_chanel</span><span class="p">):</span>
  117. <span class="c1"># getting flatten information from image and noise</span>
  118. <span class="n">image_array_chanel</span> <span class="o">=</span> <span class="n">image_array</span><span class="p">[:,</span> <span class="p">:,</span> <span class="n">chanel</span><span class="p">]</span>
  119. <span class="c1"># redefine noise if necessary</span>
  120. <span class="k">if</span> <span class="ow">not</span> <span class="n">identical</span><span class="p">:</span>
  121. <span class="n">noise_filter</span> <span class="o">=</span> <span class="n">generator</span><span class="p">(</span><span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">)</span>
  122. <span class="c1"># compute new pixel value</span>
  123. <span class="c1"># x + n * k * white_noise_filter[i] as example</span>
  124. <span class="n">noisy_image</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span>
  125. <span class="nb">list</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="n">updator</span><span class="p">,</span> <span class="n">image_array_chanel</span><span class="p">,</span> <span class="n">noise_filter</span><span class="p">)))</span>
  126. <span class="c1"># normalise values</span>
  127. <span class="n">noisy_image</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="n">_normalise</span><span class="p">,</span> <span class="n">noisy_image</span><span class="p">)),</span> <span class="s1">&#39;uint8&#39;</span><span class="p">)</span>
  128. <span class="c1"># in order to concatenate output array</span>
  129. <span class="n">noisy_image</span> <span class="o">=</span> <span class="n">noisy_image</span><span class="p">[:,</span> <span class="p">:,</span> <span class="n">np</span><span class="o">.</span><span class="n">newaxis</span><span class="p">]</span>
  130. <span class="c1"># append new chanel</span>
  131. <span class="n">output_array</span><span class="o">.</span><span class="n">append</span><span class="p">(</span><span class="n">noisy_image</span><span class="p">)</span>
  132. <span class="c1"># concatenate RGB image</span>
  133. <span class="n">output_array</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">concatenate</span><span class="p">(</span><span class="n">output_array</span><span class="p">,</span> <span class="n">axis</span><span class="o">=</span><span class="mi">2</span><span class="p">)</span>
  134. <span class="k">return</span> <span class="n">output_array</span>
  135. <div class="viewcode-block" id="white_noise"><a class="viewcode-back" href="../../../ipfml/ipfml.filters.noise.html#ipfml.filters.noise.white_noise">[docs]</a><span class="k">def</span> <span class="nf">white_noise</span><span class="p">(</span><span class="n">image</span><span class="p">,</span>
  136. <span class="n">n</span><span class="p">,</span>
  137. <span class="n">identical</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
  138. <span class="n">distribution_interval</span><span class="o">=</span><span class="p">(</span><span class="o">-</span><span class="mf">0.5</span><span class="p">,</span> <span class="mf">0.5</span><span class="p">),</span>
  139. <span class="n">k</span><span class="o">=</span><span class="mf">0.2</span><span class="p">):</span>
  140. <span class="sd">&quot;&quot;&quot;White noise filter to apply on image</span>
  141. <span class="sd"> Args:</span>
  142. <span class="sd"> image: image used as input (2D or 3D image representation)</span>
  143. <span class="sd"> n: used to set importance of noise [1, 999]</span>
  144. <span class="sd"> identical: keep or not identical noise distribution for each canal if RGB Image (default False)</span>
  145. <span class="sd"> distribution_interval: set the distribution interval of normal law distribution (default (-0.5, 0.5))</span>
  146. <span class="sd"> k: variable that specifies the amount of noise to be taken into account in the output image (default 0.2)</span>
  147. <span class="sd"> Returns:</span>
  148. <span class="sd"> 2D Numpy array with white noise applied</span>
  149. <span class="sd"> Example:</span>
  150. <span class="sd"> &gt;&gt;&gt; from ipfml.filters.noise import white_noise</span>
  151. <span class="sd"> &gt;&gt;&gt; import numpy as np</span>
  152. <span class="sd"> &gt;&gt;&gt; image = np.random.uniform(0, 255, 10000).reshape((100, 100))</span>
  153. <span class="sd"> &gt;&gt;&gt; noisy_image = white_noise(image, 10)</span>
  154. <span class="sd"> &gt;&gt;&gt; noisy_image.shape</span>
  155. <span class="sd"> (100, 100)</span>
  156. <span class="sd"> &quot;&quot;&quot;</span>
  157. <span class="n">a</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="n">distribution_interval</span>
  158. <span class="n">generator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">uniform</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">))</span>
  159. <span class="n">updator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">,</span> <span class="n">noise</span><span class="p">:</span> <span class="n">x</span> <span class="o">+</span> <span class="n">n</span> <span class="o">*</span> <span class="n">k</span> <span class="o">*</span> <span class="n">noise</span>
  160. <span class="k">return</span> <span class="n">_global_noise_filter</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">generator</span><span class="p">,</span> <span class="n">updator</span><span class="p">,</span> <span class="n">identical</span><span class="p">)</span></div>
  161. <div class="viewcode-block" id="gaussian_noise"><a class="viewcode-back" href="../../../ipfml/ipfml.filters.noise.html#ipfml.filters.noise.gaussian_noise">[docs]</a><span class="k">def</span> <span class="nf">gaussian_noise</span><span class="p">(</span><span class="n">image</span><span class="p">,</span>
  162. <span class="n">n</span><span class="p">,</span>
  163. <span class="n">identical</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
  164. <span class="n">distribution_interval</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
  165. <span class="n">k</span><span class="o">=</span><span class="mf">0.1</span><span class="p">):</span>
  166. <span class="sd">&quot;&quot;&quot;Gaussian noise filter to apply on image</span>
  167. <span class="sd"> Args:</span>
  168. <span class="sd"> image: image used as input (2D or 3D image representation)</span>
  169. <span class="sd"> n: used to set importance of noise [1, 999]</span>
  170. <span class="sd"> identical: keep or not identical noise distribution for each canal if RGB Image (default False)</span>
  171. <span class="sd"> distribution_interval: set the distribution interval of normal law distribution (default (0, 1))</span>
  172. <span class="sd"> k: variable that specifies the amount of noise to be taken into account in the output image (default 0.1)</span>
  173. <span class="sd"> Returns:</span>
  174. <span class="sd"> 2D Numpy array with gaussian noise applied</span>
  175. <span class="sd"> Example:</span>
  176. <span class="sd"> &gt;&gt;&gt; from ipfml.filters.noise import gaussian_noise</span>
  177. <span class="sd"> &gt;&gt;&gt; import numpy as np</span>
  178. <span class="sd"> &gt;&gt;&gt; image = np.random.uniform(0, 255, 10000).reshape((100, 100))</span>
  179. <span class="sd"> &gt;&gt;&gt; noisy_image = gaussian_noise(image, 10)</span>
  180. <span class="sd"> &gt;&gt;&gt; noisy_image.shape</span>
  181. <span class="sd"> (100, 100)</span>
  182. <span class="sd"> &quot;&quot;&quot;</span>
  183. <span class="n">a</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="n">distribution_interval</span>
  184. <span class="n">generator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">normal</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">))</span>
  185. <span class="n">updator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">,</span> <span class="n">noise</span><span class="p">:</span> <span class="n">x</span> <span class="o">+</span> <span class="n">n</span> <span class="o">*</span> <span class="n">k</span> <span class="o">*</span> <span class="n">noise</span>
  186. <span class="k">return</span> <span class="n">_global_noise_filter</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">generator</span><span class="p">,</span> <span class="n">updator</span><span class="p">,</span> <span class="n">identical</span><span class="p">)</span></div>
  187. <div class="viewcode-block" id="laplace_noise"><a class="viewcode-back" href="../../../ipfml/ipfml.filters.noise.html#ipfml.filters.noise.laplace_noise">[docs]</a><span class="k">def</span> <span class="nf">laplace_noise</span><span class="p">(</span><span class="n">image</span><span class="p">,</span>
  188. <span class="n">n</span><span class="p">,</span>
  189. <span class="n">identical</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
  190. <span class="n">distribution_interval</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
  191. <span class="n">k</span><span class="o">=</span><span class="mf">0.1</span><span class="p">):</span>
  192. <span class="sd">&quot;&quot;&quot;Laplace noise filter to apply on image</span>
  193. <span class="sd"> Args:</span>
  194. <span class="sd"> image: image used as input (2D or 3D image representation)</span>
  195. <span class="sd"> n: used to set importance of noise [1, 999]</span>
  196. <span class="sd"> identical: keep or not identical noise distribution for each canal if RGB Image (default False)</span>
  197. <span class="sd"> distribution_interval: set the distribution interval of normal law distribution (default (0, 1))</span>
  198. <span class="sd"> k: variable that specifies the amount of noise to be taken into account in the output image (default 0.1)</span>
  199. <span class="sd"> Returns:</span>
  200. <span class="sd"> 2D Numpay array with Laplace noise applied</span>
  201. <span class="sd"> Example:</span>
  202. <span class="sd"> &gt;&gt;&gt; from ipfml.filters.noise import laplace_noise</span>
  203. <span class="sd"> &gt;&gt;&gt; import numpy as np</span>
  204. <span class="sd"> &gt;&gt;&gt; image = np.random.uniform(0, 255, 10000).reshape((100, 100))</span>
  205. <span class="sd"> &gt;&gt;&gt; noisy_image = laplace_noise(image, 10)</span>
  206. <span class="sd"> &gt;&gt;&gt; noisy_image.shape</span>
  207. <span class="sd"> (100, 100)</span>
  208. <span class="sd"> &quot;&quot;&quot;</span>
  209. <span class="n">a</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="n">distribution_interval</span>
  210. <span class="n">generator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">laplace</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">))</span>
  211. <span class="n">updator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">,</span> <span class="n">noise</span><span class="p">:</span> <span class="n">x</span> <span class="o">+</span> <span class="n">n</span> <span class="o">*</span> <span class="n">k</span> <span class="o">*</span> <span class="n">noise</span>
  212. <span class="k">return</span> <span class="n">_global_noise_filter</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">generator</span><span class="p">,</span> <span class="n">updator</span><span class="p">,</span> <span class="n">identical</span><span class="p">)</span></div>
  213. <div class="viewcode-block" id="cauchy_noise"><a class="viewcode-back" href="../../../ipfml/ipfml.filters.noise.html#ipfml.filters.noise.cauchy_noise">[docs]</a><span class="k">def</span> <span class="nf">cauchy_noise</span><span class="p">(</span><span class="n">image</span><span class="p">,</span>
  214. <span class="n">n</span><span class="p">,</span>
  215. <span class="n">identical</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
  216. <span class="n">distribution_interval</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
  217. <span class="n">k</span><span class="o">=</span><span class="mf">0.0002</span><span class="p">):</span>
  218. <span class="sd">&quot;&quot;&quot;Cauchy noise filter to apply on image</span>
  219. <span class="sd"> Args:</span>
  220. <span class="sd"> image: image used as input (2D or 3D image representation)</span>
  221. <span class="sd"> n: used to set importance of noise [1, 999]</span>
  222. <span class="sd"> identical: keep or not identical noise distribution for each canal if RGB Image (default False)</span>
  223. <span class="sd"> distribution_interval: set the distribution interval of normal law distribution (default (0, 1))</span>
  224. <span class="sd"> k: variable that specifies the amount of noise to be taken into account in the output image (default 0.0002)</span>
  225. <span class="sd"> Returns:</span>
  226. <span class="sd"> 2D Numpy array with Cauchy noise applied</span>
  227. <span class="sd"> Example:</span>
  228. <span class="sd"> &gt;&gt;&gt; from ipfml.filters.noise import cauchy_noise</span>
  229. <span class="sd"> &gt;&gt;&gt; import numpy as np</span>
  230. <span class="sd"> &gt;&gt;&gt; image = np.random.uniform(0, 255, 10000).reshape((100, 100))</span>
  231. <span class="sd"> &gt;&gt;&gt; noisy_image = cauchy_noise(image, 10)</span>
  232. <span class="sd"> &gt;&gt;&gt; noisy_image.shape</span>
  233. <span class="sd"> (100, 100)</span>
  234. <span class="sd"> &quot;&quot;&quot;</span>
  235. <span class="n">a</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="n">distribution_interval</span>
  236. <span class="n">generator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">standard_cauchy</span><span class="p">((</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">))</span>
  237. <span class="n">updator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">,</span> <span class="n">noise</span><span class="p">:</span> <span class="n">x</span> <span class="o">+</span> <span class="n">n</span> <span class="o">*</span> <span class="n">k</span> <span class="o">*</span> <span class="n">noise</span>
  238. <span class="k">return</span> <span class="n">_global_noise_filter</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">generator</span><span class="p">,</span> <span class="n">updator</span><span class="p">,</span> <span class="n">identical</span><span class="p">)</span></div>
  239. <div class="viewcode-block" id="log_normal_noise"><a class="viewcode-back" href="../../../ipfml/ipfml.filters.noise.html#ipfml.filters.noise.log_normal_noise">[docs]</a><span class="k">def</span> <span class="nf">log_normal_noise</span><span class="p">(</span><span class="n">image</span><span class="p">,</span>
  240. <span class="n">n</span><span class="p">,</span>
  241. <span class="n">identical</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
  242. <span class="n">distribution_interval</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
  243. <span class="n">k</span><span class="o">=</span><span class="mf">0.05</span><span class="p">):</span>
  244. <span class="sd">&quot;&quot;&quot;Log-normal noise filter to apply on image</span>
  245. <span class="sd"> Args:</span>
  246. <span class="sd"> image: image used as input (2D or 3D image representation)</span>
  247. <span class="sd"> n: used to set importance of noise [1, 999]</span>
  248. <span class="sd"> identical: keep or not identical noise distribution for each canal if RGB Image (default False)</span>
  249. <span class="sd"> distribution_interval: set the distribution interval of normal law distribution (default (0, 1))</span>
  250. <span class="sd"> k: variable that specifies the amount of noise to be taken into account in the output image (default 0.05)</span>
  251. <span class="sd"> Returns:</span>
  252. <span class="sd"> 2D Numpy array with Log-normal noise applied</span>
  253. <span class="sd"> Example:</span>
  254. <span class="sd"> &gt;&gt;&gt; from ipfml.filters.noise import log_normal_noise</span>
  255. <span class="sd"> &gt;&gt;&gt; import numpy as np</span>
  256. <span class="sd"> &gt;&gt;&gt; image = np.random.uniform(0, 255, 10000).reshape((100, 100))</span>
  257. <span class="sd"> &gt;&gt;&gt; noisy_image = log_normal_noise(image, 10)</span>
  258. <span class="sd"> &gt;&gt;&gt; noisy_image.shape</span>
  259. <span class="sd"> (100, 100)</span>
  260. <span class="sd"> &quot;&quot;&quot;</span>
  261. <span class="n">a</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="n">distribution_interval</span>
  262. <span class="n">generator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">lognormal</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">))</span>
  263. <span class="n">updator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">,</span> <span class="n">noise</span><span class="p">:</span> <span class="n">x</span> <span class="o">+</span> <span class="n">n</span> <span class="o">*</span> <span class="n">k</span> <span class="o">*</span> <span class="n">noise</span>
  264. <span class="k">return</span> <span class="n">_global_noise_filter</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">generator</span><span class="p">,</span> <span class="n">updator</span><span class="p">,</span> <span class="n">identical</span><span class="p">)</span></div>
  265. <div class="viewcode-block" id="mut_white_noise"><a class="viewcode-back" href="../../../ipfml/ipfml.filters.noise.html#ipfml.filters.noise.mut_white_noise">[docs]</a><span class="k">def</span> <span class="nf">mut_white_noise</span><span class="p">(</span><span class="n">image</span><span class="p">,</span>
  266. <span class="n">n</span><span class="p">,</span>
  267. <span class="n">identical</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span>
  268. <span class="n">distribution_interval</span><span class="o">=</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">),</span>
  269. <span class="n">k</span><span class="o">=</span><span class="mf">0.002</span><span class="p">):</span>
  270. <span class="sd">&quot;&quot;&quot;Multiplied White noise filter to apply on image</span>
  271. <span class="sd"> Args:</span>
  272. <span class="sd"> image: image used as input (2D or 3D image representation)</span>
  273. <span class="sd"> n: used to set importance of noise [1, 999]</span>
  274. <span class="sd"> identical: keep or not identical noise distribution for each canal if RGB Image (default False)</span>
  275. <span class="sd"> distribution_interval: set the distribution interval of normal law distribution (default (0, 1))</span>
  276. <span class="sd"> k: variable that specifies the amount of noise to be taken into account in the output image (default 0.002)</span>
  277. <span class="sd"> Returns:</span>
  278. <span class="sd"> 2D Numpy array with multiplied white noise applied</span>
  279. <span class="sd"> Example:</span>
  280. <span class="sd"> &gt;&gt;&gt; from ipfml.filters.noise import mut_white_noise</span>
  281. <span class="sd"> &gt;&gt;&gt; import numpy as np</span>
  282. <span class="sd"> &gt;&gt;&gt; image = np.random.uniform(0, 255, 10000).reshape((100, 100))</span>
  283. <span class="sd"> &gt;&gt;&gt; noisy_image = mut_white_noise(image, 10)</span>
  284. <span class="sd"> &gt;&gt;&gt; noisy_image.shape</span>
  285. <span class="sd"> (100, 100)</span>
  286. <span class="sd"> &quot;&quot;&quot;</span>
  287. <span class="n">min_value</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">-</span> <span class="p">(</span><span class="n">k</span> <span class="o">/</span> <span class="mi">2</span><span class="p">)</span>
  288. <span class="n">max_value</span> <span class="o">=</span> <span class="mi">1</span> <span class="o">+</span> <span class="p">(</span><span class="n">k</span> <span class="o">/</span> <span class="mi">2</span><span class="p">)</span>
  289. <span class="n">a</span><span class="p">,</span> <span class="n">b</span> <span class="o">=</span> <span class="n">distribution_interval</span>
  290. <span class="n">generator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">:</span> <span class="n">min_value</span> <span class="o">+</span> <span class="p">(</span><span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">uniform</span><span class="p">(</span><span class="n">a</span><span class="p">,</span> <span class="n">b</span><span class="p">,</span> <span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">))</span> <span class="o">*</span> <span class="p">(</span><span class="n">max_value</span> <span class="o">-</span> <span class="n">min_value</span><span class="p">))</span>
  291. <span class="n">updator</span> <span class="o">=</span> <span class="k">lambda</span> <span class="n">x</span><span class="p">,</span> <span class="n">noise</span><span class="p">:</span> <span class="n">x</span> <span class="o">*</span> <span class="nb">pow</span><span class="p">(</span><span class="n">noise</span><span class="p">,</span> <span class="n">n</span><span class="p">)</span>
  292. <span class="k">return</span> <span class="n">_global_noise_filter</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">generator</span><span class="p">,</span> <span class="n">updator</span><span class="p">,</span> <span class="n">identical</span><span class="p">)</span></div>
  293. <div class="viewcode-block" id="salt_pepper_noise"><a class="viewcode-back" href="../../../ipfml/ipfml.filters.noise.html#ipfml.filters.noise.salt_pepper_noise">[docs]</a><span class="k">def</span> <span class="nf">salt_pepper_noise</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">n</span><span class="p">,</span> <span class="n">identical</span><span class="o">=</span><span class="kc">False</span><span class="p">,</span> <span class="n">p</span><span class="o">=</span><span class="mf">0.1</span><span class="p">,</span> <span class="n">k</span><span class="o">=</span><span class="mf">0.5</span><span class="p">):</span>
  294. <span class="sd">&quot;&quot;&quot;Pepper salt noise filter to apply on image</span>
  295. <span class="sd"> Args:</span>
  296. <span class="sd"> image: image used as input (2D or 3D image representation)</span>
  297. <span class="sd"> n: used to set importance of noise [1, 999]</span>
  298. <span class="sd"> identical: keep or not identical noise distribution for each canal if RGB Image (default False)</span>
  299. <span class="sd"> p: probability to increase pixel value otherwise decrease it</span>
  300. <span class="sd"> k: variable that specifies the amount of noise to be taken into account in the output image (default 0.5)</span>
  301. <span class="sd"> Returns:</span>
  302. <span class="sd"> 2D Numpy array with salt and pepper noise applied</span>
  303. <span class="sd"> Example:</span>
  304. <span class="sd"> &gt;&gt;&gt; from ipfml.filters.noise import salt_pepper_noise</span>
  305. <span class="sd"> &gt;&gt;&gt; import numpy as np</span>
  306. <span class="sd"> &gt;&gt;&gt; image = np.random.uniform(0, 255, 10000).reshape((100, 100))</span>
  307. <span class="sd"> &gt;&gt;&gt; noisy_image = salt_pepper_noise(image, 10)</span>
  308. <span class="sd"> &gt;&gt;&gt; noisy_image.shape</span>
  309. <span class="sd"> (100, 100)</span>
  310. <span class="sd"> &quot;&quot;&quot;</span>
  311. <span class="k">def</span> <span class="nf">_generator</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">):</span>
  312. <span class="n">x</span> <span class="o">=</span> <span class="n">w</span> <span class="o">*</span> <span class="n">h</span>
  313. <span class="n">nb_elem</span> <span class="o">=</span> <span class="nb">int</span><span class="p">(</span><span class="n">p</span> <span class="o">*</span> <span class="n">x</span><span class="p">)</span>
  314. <span class="n">elements</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">full</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="mi">0</span><span class="p">)</span>
  315. <span class="n">elements</span><span class="p">[</span><span class="mi">0</span><span class="p">:</span><span class="n">nb_elem</span><span class="p">]</span> <span class="o">=</span> <span class="mi">1</span>
  316. <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">shuffle</span><span class="p">(</span><span class="n">elements</span><span class="p">)</span>
  317. <span class="k">return</span> <span class="n">elements</span><span class="o">.</span><span class="n">reshape</span><span class="p">(</span><span class="n">h</span><span class="p">,</span> <span class="n">w</span><span class="p">)</span>
  318. <span class="n">image_array</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="n">image</span><span class="p">)</span>
  319. <span class="k">if</span> <span class="n">image_array</span><span class="o">.</span><span class="n">ndim</span> <span class="o">==</span> <span class="mi">3</span><span class="p">:</span>
  320. <span class="n">height</span><span class="p">,</span> <span class="n">width</span><span class="p">,</span> <span class="n">_</span> <span class="o">=</span> <span class="n">image_array</span><span class="o">.</span><span class="n">shape</span>
  321. <span class="k">else</span><span class="p">:</span>
  322. <span class="n">height</span><span class="p">,</span> <span class="n">width</span> <span class="o">=</span> <span class="n">image_array</span><span class="o">.</span><span class="n">shape</span>
  323. <span class="c1"># need same random variable for each pixel value if identical</span>
  324. <span class="k">if</span> <span class="n">identical</span><span class="p">:</span>
  325. <span class="n">gen</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">random</span><span class="o">.</span><span class="n">uniform</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">,</span> <span class="n">height</span> <span class="o">*</span> <span class="n">width</span><span class="p">)</span>
  326. <span class="k">for</span> <span class="n">i</span> <span class="ow">in</span> <span class="nb">range</span><span class="p">(</span><span class="mi">2</span><span class="p">):</span>
  327. <span class="n">gen</span> <span class="o">=</span> <span class="n">np</span><span class="o">.</span><span class="n">insert</span><span class="p">(</span><span class="n">gen</span><span class="p">,</span> <span class="mi">0</span><span class="p">,</span> <span class="n">gen</span><span class="p">)</span>
  328. <span class="n">gen</span> <span class="o">=</span> <span class="nb">iter</span><span class="p">(</span><span class="n">gen</span><span class="p">)</span>
  329. <span class="c1"># here noise variable is boolean to update or not pixel value</span>
  330. <span class="k">def</span> <span class="nf">_updator</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">noise</span><span class="p">):</span>
  331. <span class="c1"># apply specific changes to each value of 1D array</span>
  332. <span class="k">if</span> <span class="nb">isinstance</span><span class="p">(</span><span class="n">x</span><span class="p">,</span> <span class="n">np</span><span class="o">.</span><span class="n">ndarray</span><span class="p">):</span>
  333. <span class="k">return</span> <span class="n">np</span><span class="o">.</span><span class="n">array</span><span class="p">(</span><span class="nb">list</span><span class="p">(</span><span class="nb">map</span><span class="p">(</span><span class="n">_updator</span><span class="p">,</span> <span class="n">x</span><span class="p">,</span> <span class="n">noise</span><span class="p">)))</span>
  334. <span class="c1"># probabilty to increase or decrease pixel value</span>
  335. <span class="k">if</span> <span class="n">identical</span><span class="p">:</span>
  336. <span class="n">rand</span> <span class="o">=</span> <span class="nb">next</span><span class="p">(</span><span class="n">gen</span><span class="p">)</span>
  337. <span class="k">else</span><span class="p">:</span>
  338. <span class="n">rand</span> <span class="o">=</span> <span class="n">random</span><span class="o">.</span><span class="n">uniform</span><span class="p">(</span><span class="mi">0</span><span class="p">,</span> <span class="mi">1</span><span class="p">)</span>
  339. <span class="k">if</span> <span class="n">noise</span><span class="p">:</span>
  340. <span class="k">if</span> <span class="n">rand</span> <span class="o">&gt;</span> <span class="mf">0.5</span><span class="p">:</span>
  341. <span class="k">return</span> <span class="n">x</span> <span class="o">+</span> <span class="n">n</span> <span class="o">*</span> <span class="n">k</span>
  342. <span class="k">else</span><span class="p">:</span>
  343. <span class="k">return</span> <span class="n">x</span> <span class="o">-</span> <span class="n">n</span> <span class="o">*</span> <span class="n">k</span>
  344. <span class="k">else</span><span class="p">:</span>
  345. <span class="k">return</span> <span class="n">x</span>
  346. <span class="k">return</span> <span class="n">_global_noise_filter</span><span class="p">(</span><span class="n">image</span><span class="p">,</span> <span class="n">_generator</span><span class="p">,</span> <span class="n">_updator</span><span class="p">,</span> <span class="n">identical</span><span class="p">)</span></div>
  347. </pre></div>
  348. </div>
  349. </div>
  350. <footer>
  351. <hr/>
  352. <div role="contentinfo">
  353. <p>
  354. &copy; Copyright 2019, Jérôme BUISINE
  355. </p>
  356. </div>
  357. 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>.
  358. </footer>
  359. </div>
  360. </div>
  361. </section>
  362. </div>
  363. <script type="text/javascript" id="documentation_options" data-url_root="../../../" src="../../../_static/documentation_options.js"></script>
  364. <script type="text/javascript" src="../../../_static/jquery.js"></script>
  365. <script type="text/javascript" src="../../../_static/underscore.js"></script>
  366. <script type="text/javascript" src="../../../_static/doctools.js"></script>
  367. <script type="text/javascript" src="../../../_static/language_data.js"></script>
  368. <script type="text/javascript" src="../../../_static/js/theme.js"></script>
  369. <script type="text/javascript">
  370. jQuery(function () {
  371. SphinxRtdTheme.Navigation.enable(true);
  372. });
  373. </script>
  374. </body>
  375. </html>