MATLAB--形态学图像处理

谁借莪１个温暖的怀抱￠ 2022-05-11 01:40 586阅读 0赞

#### **腐蚀** ####

##### **imerode(image, s) 对图像进行腐蚀操作** #####

f = imread('img8.tif');
    se = strel('disk', 10);
    g = imerode(f, se);
    subplot(1, 2, 1), imshow(f), title('原图');
    subplot(1, 2, 2), imshow(g), title('腐蚀后');
     
      
       1 
       2 
       3 
       4 
       5

利用一个10\*10的菱形进行腐蚀  
\* 输入:  
![这里写图片描述][SouthEast]  
![这里写图片描述][SouthEast 1]  
利用其它大小的菱形进行腐蚀后的结果

f = imread('img8.tif');
    g1 = imerode(f, strel('disk', 5));
    g2 = imerode(f, strel('disk', 10));
    g3 = imerode(f, strel('disk', 20));
    subplot(2, 2, 1), imshow(f), title('原图');
    subplot(2, 2, 2), imshow(g1), title('5*5 腐蚀后');
    subplot(2, 2, 3), imshow(g2), title('10*10 腐蚀后');
    subplot(2, 2, 4), imshow(g3), title('20*20 腐蚀后');
     
      
       1 
       2 
       3 
       4 
       5 
       6 
       7 
       8

*  输入:  
    ![这里写图片描述][SouthEast 2]
 *  输出:  
    ![这里写图片描述][SouthEast 3]

#### **腐蚀和膨胀的组合** ####

##### **imopen(A, B) 将A，B进行开运算** #####

A被B腐蚀后再用B来膨胀

f = imread('img9.tif');
    se = strel('square', 20);
    g = imopen(f, se);
    g1 = imerode(f, se);
    g2 = imdilate(g1, se);
    subplot(2, 2, 1), imshow(f), title('原图');
    subplot(2, 2, 2), imshow(g1), title('腐蚀运算');
    subplot(2, 2, 3), imshow(g2), title('膨胀运算(开运算)');
    subplot(2, 2, 4), imshow(g), title('开运算');
     
      
       1 
       2 
       3 
       4 
       5 
       6 
       7 
       8 
       9

*  输入:  
    ![这里写图片描述][SouthEast 4]
 *  输出:  
    ![这里写图片描述][SouthEast 5]

##### **imclose(A, B) 将A，B进行闭运算** #####

A被B膨胀在用B腐蚀，跟开运算相反

f = imread('img9.tif');
    se = strel('square', 20);
    g = imclose(f, se);
    g1 = imdilate(f, se);
    g2 = imerode(g1, se);
    subplot(2, 2, 1), imshow(f), title('原图');
    subplot(2, 2, 2), imshow(g1), title('膨胀运算');
    subplot(2, 2, 3), imshow(g2), title('腐蚀运算(闭运算)');
    subplot(2, 2, 4), imshow(g), title('闭运算');
     
      
       1 
       2 
       3 
       4 
       5 
       6 
       7 
       8 
       9

*  输入:  
    ![这里写图片描述][SouthEast 6]
 *  输出:  
    ![这里写图片描述][SouthEast 7]

一般来说都是将开闭操作配合使用，可以达到去除噪声的效果。  
在这之前，我们先研究一下开闭操作和闭开操作

f = imread('img9.tif');
    se = strel('square', 20);
    g1 = imopen(f, se);
    g2 = imclose(f, se);
    g3 = imopen(g2, se);
    g4 = imclose(g1, se);
    subplot(2, 2, 1), imshow(g1), title('开操作');
    subplot(2, 2, 2), imshow(g2), title('闭操作');
    subplot(2, 2, 3), imshow(g3), title('闭开操作');
    subplot(2, 2, 4), imshow(g4), title('开闭操作');
     
      
       1 
       2 
       3 
       4 
       5 
       6 
       7 
       8 
       9 
       10

*  输入:  
    ![这里写图片描述][SouthEast 8]
 *  输出:  
    ![这里写图片描述][SouthEast 9]

可以看出，闭开 和 开闭是没有区别的!所以先开或者先闭得到的结果也就一样啦。  
接着看看他去除噪音的效果

f = imread('img10.tif');
    se = strel('square', 3);
    g1 = imopen(f, se);
    g2 = imclose(f, se);
    g3 = imclose(g1, se);
    subplot(2, 2, 2), imshow(g1), title('开操作');
    subplot(2, 2, 3), imshow(g2), title('闭操作');
    subplot(2, 2, 4), imshow(g3), title('开闭操作');
    subplot(2, 2, 1), imshow(f), title('原图');
     
      
       1 
       2 
       3 
       4 
       5 
       6 
       7 
       8 
       9

*  输入:  
    ![这里写图片描述][SouthEast 10]
 *  输出:  
    ![这里写图片描述][SouthEast 11]

##### **bwhitmiss(A, B1, B2) 击中或未击中, B1为击中的结构，B2为未击中的结构** #####

取B1击中的地方和B2未击中的地方

f = imread('img11.tif');
    B1 = [0 0 0; 0 1 1; 0 1 0];
    B2 = [1 1 1; 1 0 0; 1 0 0];
    g = bwhitmiss(f, B1, B2);
    subplot(1, 2, 1), imshow(f), title('原图');
    subplot(1, 2, 2), imshow(g), title('处理后');
     
      
       1 
       2 
       3 
       4 
       5 
       6

*  输入:  
    ![这里写图片描述][SouthEast 12]
 *  输出:  
    ![这里写图片描述][SouthEast 13]

##### \*\*endpoints(f) 使用makelut 和 applylut 在二值图中检测端点 #####

f = imread('img12.tif');
    g = endpoints(f);
    subplot(1, 2, 1), imshow(f), title('原图');
    subplot(1, 2, 2), imshow(g), title('端点');
     
      
       1 
       2 
       3 
       4

*  输入:  
    ![这里写图片描述][SouthEast 14]
 *  输出:  
    ![这里写图片描述][SouthEast 15]  
    书中大脸猫实验  
    `lut = makelut(@conwaylaws, 3); bw1 = [0 0 0 0 0 0 0 0 0 0; 0 0 0 0 0 0 0 0 0 0; 0 0 0 1 0 0 1 0 0 0; 0 0 0 1 1 1 1 0 0 0; 0 0 1 0 0 0 0 1 0 0; 0 0 1 0 0 0 0 1 0 0; 0 0 1 0 0 0 0 1 0 0; 0 0 0 1 1 1 1 0 0 0; 0 0 0 0 0 0 0 0 0 0; 0 0 0 0 0 0 0 0 0 0; ]; subplot(2, 2, 1), imshow(bw1), title('1'); bw2 = applylut(bw1, lut); subplot(2, 2, 2), imshow(bw2), title('2'); bw3 = applylut(bw2, lut); subplot(2, 2, 3), imshow(bw3), title('3'); bw4 = applylut(bw3, lut); subplot(2, 2, 4), imshow(bw4), title('4');`  
    
    
     *  输入:  
        ![这里写图片描述][SouthEast 16]
 *  输出:  
    ![这里写图片描述][SouthEast 17]

##### **bwmorph(f, opeartion, n) 将操作（膨胀，腐蚀，查找）进行组合操作** #####

![这里写图片描述][SouthEast 18]

细化操作 thin

f = imread('img10.tif');
    f = bwmorph(f, 'close', 1);
    f = bwmorph(f, 'open', 1);
    g1 = bwmorph(f, 'thin', 1);
    g2 = bwmorph(f, 'thin', 2);
    g3 = bwmorph(f, 'thin', Inf);
    subplot(2, 2, 1), imshow(f), title('原图');
    subplot(2, 2, 2), imshow(g1), title('1');
    subplot(2, 2, 3), imshow(g2), title('2');
    subplot(2, 2, 4), imshow(g3), title('Inf');
     
      
       1 
       2 
       3 
       4 
       5 
       6 
       7 
       8 
       9 
       10

*  输入:  
    ![这里写图片描述][SouthEast 19]
 *  输出:  
    ![这里写图片描述][SouthEast 20]

骨骼操作 skel

f = imread('img13.tif');
    g1 = bwmorph(f, 'skel', Inf);
    g2 = g1;
    for k = 1 : 5
        g2 = g2 & ~endpoints(g2);
    end
    subplot(2, 2, 1), imshow(f), title('原图');
    subplot(2, 2, 2), imshow(g1), title('skel');
    subplot(2, 2, 3), imshow(g2), title('endpoints');
     
      
       1 
       2 
       3 
       4 
       5 
       6 
       7 
       8 
       9

*  输入:  
    ![这里写图片描述][SouthEast 21]
 *  输出:  
    ![这里写图片描述][SouthEast 22]

[SouthEast]: /images/20220511/979ba31fc21a45f69fd7d91f14c969e4.png
[SouthEast 1]: /images/20220511/7de715cf25e94704b6f8d5a9b6ec1b15.png
[SouthEast 2]: /images/20220511/79ad784aa53f46a9b077384ab702eebc.png
[SouthEast 3]: /images/20220511/2f4f365897124772a3ca3a0d071dc319.png
[SouthEast 4]: /images/20220511/05f0a8cca769490bbf0f2e5b1eb887ae.png
[SouthEast 5]: /images/20220511/e152bb78e6c84c6aa7aa22903163659e.png
[SouthEast 6]: /images/20220511/662bc4050a8344c2b631d6fc10d46019.png
[SouthEast 7]: /images/20220511/4270195baee448ee9b3681c064541113.png
[SouthEast 8]: /images/20220511/f34a93e03a1143eb9c413fd81f8f265b.png
[SouthEast 9]: /images/20220511/7c16c84f5812409cbcc47b83993d4266.png
[SouthEast 10]: /images/20220511/b95ed10cf4404dfc80e1e469e50bfdb1.png
[SouthEast 11]: /images/20220511/5a743d5b0a054816872cc293d18a9bed.png
[SouthEast 12]: /images/20220511/1fc40762957e4cfc929d49f5ecce7cf6.png
[SouthEast 13]: /images/20220511/405847213e70409083c75ce5189ba206.png
[SouthEast 14]: /images/20220511/5c8d74829d664c5cb3d38ed55261209e.png
[SouthEast 15]: /images/20220511/69bccd87ff9643848aa646404d146aa9.png
[SouthEast 16]: /images/20220511/a9f08a312fca4222a71ca7595059c496.png
[SouthEast 17]: /images/20220511/c29072e48f634945bd363dafebd43742.png
[SouthEast 18]: /images/20220511/5fa1efe6072e4bf2be258ed4bf18a533.png
[SouthEast 19]: /images/20220511/6a6cced7ebbc452fb9dd5bb164d181e1.png
[SouthEast 20]: /images/20220511/ed14bb932c114413a7d08bb6fc5e8c31.png
[SouthEast 21]: /images/20220511/a9819071b0c74291b774186a112508b0.png
[SouthEast 22]: /images/20220511/299c1d346bac4fd99a97fe200b0f281c.png