opencv笔记（二）——图像的膨胀（dilate）和腐蚀（erode）

冷不防 2022-05-19 12:50 583阅读 0赞

# 一、图像腐蚀（erode）： #

**1.1 官方文档说明：**

<table> 
 <tbody> 
  <tr> 
   <td>void cv::erode</td> 
   <td>(</td> 
   <td><a href="https://docs.opencv.org/3.4.1/dc/d84/group__core__basic.html#gab1371a23c90d5f4725ea98240377d7a1" rel="nofollow">InputArray</a>&nbsp;</td> 
   <td><em>src</em>,</td> 
  </tr> 
  <tr> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
   <td><a href="https://docs.opencv.org/3.4.1/dc/d84/group__core__basic.html#gaad17fda1d0f0d1ee069aebb1df2913c0" rel="nofollow">OutputArray</a>&nbsp;</td> 
   <td><em>dst</em>,</td> 
  </tr> 
  <tr> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
   <td><a href="https://docs.opencv.org/3.4.1/dc/d84/group__core__basic.html#gab1371a23c90d5f4725ea98240377d7a1" rel="nofollow">InputArray</a>&nbsp;</td> 
   <td><em>kernel</em>,</td> 
  </tr> 
  <tr> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
   <td><a href="https://docs.opencv.org/3.4.1/dc/d84/group__core__basic.html#ga1e83eafb2d26b3c93f09e8338bcab192" rel="nofollow">Point</a>&nbsp;</td> 
   <td><em>anchor</em>&nbsp;=&nbsp;<code><a href="https://docs.opencv.org/3.4.1/dc/d84/group__core__basic.html#ga1e83eafb2d26b3c93f09e8338bcab192" rel="nofollow">Point</a>(-1,-1)</code>,</td> 
  </tr> 
  <tr> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
   <td>int&nbsp;</td> 
   <td><em>iterations</em>&nbsp;=&nbsp;<code>1</code>,</td> 
  </tr> 
  <tr> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
   <td>int&nbsp;</td> 
   <td><em>borderType</em>&nbsp;=&nbsp;<code><a href="https://docs.opencv.org/3.4.1/d2/de8/group__core__array.html#gga209f2f4869e304c82d07739337eae7c5aed2e4346047e265c8c5a6d0276dcd838" rel="nofollow">BORDER_CONSTANT</a></code>,</td> 
  </tr> 
  <tr> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
   <td>const&nbsp;<a href="https://docs.opencv.org/3.4.1/dc/d84/group__core__basic.html#ga599fe92e910c027be274233eccad7beb" rel="nofollow">Scalar</a>&nbsp;&amp;&nbsp;</td> 
   <td><em>borderValue</em>&nbsp;=&nbsp;<code><a href="https://docs.opencv.org/3.4.1/d4/d86/group__imgproc__filter.html#ga94756fad83d9d24d29c9bf478558c40a" rel="nofollow">morphologyDefaultBorderValue</a>()</code>&nbsp;</td> 
  </tr> 
  <tr> 
   <td>&nbsp;</td> 
   <td>)</td> 
   <td>&nbsp;</td> 
   <td>&nbsp;</td> 
  </tr> 
 </tbody> 
</table>

<table> 
 <tbody> 
  <tr> 
   <th colspan="999">Python:</th> 
  </tr> 
  <tr> 
   <td>&nbsp;</td> 
   <td>dst</td> 
   <td>=</td> 
   <td>cv.dilate(</td> 
   <td>src, kernel[, dst[, anchor[, iterations[, borderType[, borderValue]]]]]</td> 
   <td>)</td> 
  </tr> 
 </tbody> 
</table>

### 1.2 参数说明 ###

<table> 
 <tbody> 
  <tr> 
   <td>src</td> 
   <td>输入的图像名字；信道的数目可以是任意的，但深度应该是cv_8u、cv_16u、cv_16、cv_32f或cv_64f。</td> 
  </tr> 
  <tr> 
   <td>dst</td> 
   <td>输出与输入相同大小和类型的图像.</td> 
  </tr> 
  <tr> 
   <td>kernel</td> 
   <td>用于侵蚀的结构元素；如果元素=mat()，则使用3 x 3矩形结构元素。可以使用<span style="color:#e579b6;"><code>getStructuringElement</code></span>来创建结构元素。。</td> 
  </tr> 
  <tr> 
   <td>anchor</td> 
   <td>结构元素的锚点位置，默认值value(-1,-1)表示锚点位于结构元素中心</td> 
  </tr> 
  <tr> 
   <td>iterations</td> 
   <td>腐蚀操作被递归执行的次数</td> 
  </tr> 
  <tr> 
   <td>borderType</td> 
   <td>推断边缘类型，可参考<a href="https://docs.opencv.org/3.4.1/d2/de8/group__core__array.html#ga209f2f4869e304c82d07739337eae7c5" rel="nofollow">BorderTypes</a></td> 
  </tr> 
  <tr> 
   <td>borderValue</td> 
   <td>在边框不变的情况下的边界值</td> 
  </tr> 
 </tbody> 
</table>

### 1.3 示例 ###

<table style="width:500px;"> 
 <tbody> 
  <tr> 
   <td> <pre>Mat element=getStructuringElement(MORPH_RECT, Size(5,5)); // 定义描述子形状、大小
cvtColor(Img, gray, COLOR_RGB2GRAY ); //将图像转化为灰度图像
erode(gray, gray, element);//进行腐蚀</pre> </td> 
  </tr> 
 </tbody> 
</table>

# 二、图像膨胀（dilate）： #

<table style="width:500px;"> 
 <tbody> 
  <tr> 
   <td> <pre>Mat element=getStructuringElement(MORPH_RECT, Size(5,5)); // 定义描述子形状、大小
cvtColor(Img, gray, COLOR_RGB2GRAY ); //将图像转化为灰度图像
dilate(gray, gray, element);//进行膨胀
</pre> </td> 
  </tr> 
 </tbody> 
</table>