Opencv处理图像--细化

秒速五厘米 2022-05-17 05:38 536阅读 0赞

Opencv处理图像—细化

细化的算法有很多种，但比较常用的算法是查表法

细化是从原来的图中去掉一些点，但仍要保持原来的形状。

实际上是保持原图的骨架。

代码如上：

  1. #include<iostream>
  2. #include<opencv2/core/core.hpp>
  3. #include<opencv2/highgui/highgui.hpp>
  6. IplImage* ColorSrc ,*ColorSrcCopy;
  8. //#define SHOW
  10. //#define LONG int
  11. //#define BYTE unsigned char
  12. /
  13. //基于索引表的细化细化算法
  14. //功能：对图象进行细化
  15. //参数：lpDIBBits：代表图象的一维数组
  16. //      lWidth：图象高度
  17. //      lHeight：图象宽度
  18. //      无返回值
  19. bool ThiningDIBSkeleton (unsigned char* lpDIBBits, int lWidth, int lHeight)
  20. {    
  21.     //循环变量
  22.     long i;
  23.     long j;
  24.     long lLength;
  26.     unsigned char deletemark[256] = {
  27.         0,0,0,0,0,0,0,1,    0,0,1,1,0,0,1,1,
  28.         0,0,0,0,0,0,0,0,    0,0,1,1,1,0,1,1,
  29.         0,0,0,0,0,0,0,0,    1,0,0,0,1,0,1,1,
  30.         0,0,0,0,0,0,0,0,    1,0,1,1,1,0,1,1,
  31.         0,0,0,0,0,0,0,0,    0,0,0,0,0,0,0,0,
  32.         0,0,0,0,0,0,0,0,    0,0,0,0,0,0,0,0,
  33.         0,0,0,0,0,0,0,0,    1,0,0,0,1,0,1,1,
  34.         1,0,0,0,0,0,0,0,    1,0,1,1,1,0,1,1,
  35.         0,0,1,1,0,0,1,1,    0,0,0,1,0,0,1,1,
  36.         0,0,0,0,0,0,0,0,    0,0,0,1,0,0,1,1,
  37.         1,1,0,1,0,0,0,1,    0,0,0,0,0,0,0,0,
  38.         1,1,0,1,0,0,0,1,    1,1,0,0,1,0,0,0,
  39.         0,1,1,1,0,0,1,1,    0,0,0,1,0,0,1,1,
  40.         0,0,0,0,0,0,0,0,    0,0,0,0,0,1,1,1,
  41.         1,1,1,1,0,0,1,1,    1,1,0,0,1,1,0,0,
  42.         1,1,1,1,0,0,1,1,    1,1,0,0,1,1,0,0
  43.     };//索引表
  45.     unsigned char p0, p1, p2, p3, p4, p5, p6, p7;
  46.     unsigned char *pmid, *pmidtemp;
  47.     unsigned char sum;
  48.     int changed;
  49.     bool bStart = true;
  50.     lLength = lWidth * lHeight;
  51.     unsigned char *pTemp = (unsigned char *)malloc(sizeof(unsigned char) * lWidth * lHeight);
  53.     //    P0 P1 P2
  54.     //    P7    P3
  55.     //    P6 P5 P4
  57.     while(bStart)
  58.     {
  59.         bStart = false;
  60.         changed = 0;
  62.         //首先求边缘点(并行)
  63.         pmid = (unsigned char *)lpDIBBits + lWidth + 1;
  64.         memset(pTemp,  0, lLength);
  65.         pmidtemp = (unsigned char *)pTemp + lWidth + 1;
  66.         for(i = 1; i < lHeight -1; i++)
  67.         {
  68.             for(j = 1; j < lWidth - 1; j++)
  69.             {
  70.                 if( *pmid == 0)
  71.                 {
  72.                     pmid++;
  73.                     pmidtemp++;
  74.                     continue;
  75.                 }
  77.                 p3 = *(pmid + 1);
  78.                 p2 = *(pmid + 1 - lWidth);
  79.                 p1 = *(pmid - lWidth);
  80.                 p0 = *(pmid - lWidth -1);
  81.                 p7 = *(pmid - 1);
  82.                 p6 = *(pmid + lWidth - 1);
  83.                 p5 = *(pmid + lWidth);
  84.                 p4 = *(pmid + lWidth + 1);
  86.                 sum = p0 & p1 & p2 & p3 & p4 & p5 & p6 & p7;
  87.                 if(sum == 0)
  88.                 {
  89.                     *pmidtemp = 1;
  91.                     #ifdef SHOW
  92.                     cvSet2D(ColorSrc,i,j,cvScalar(0,0,255));
  93.                     #endif
  94.                 }
  96.                 pmid++;
  97.                 pmidtemp++;
  98.             }
  99.             pmid++;
  100.             pmid++;
  101.             pmidtemp++;
  102.             pmidtemp++;
  103.         }
  104.         #ifdef SHOW
  105.         cvNamedWindow("color");
  106.         cvShowImage("color",ColorSrc);
  107.         cvWaitKey(0);
  108.         #endif
  110.         //现在开始串行删除
  111.         pmid = (unsigned char *)lpDIBBits + lWidth + 1;
  112.         pmidtemp = (unsigned char *)pTemp + lWidth + 1;
  114.         for(i = 1; i < lHeight -1; i++)
  115.         {
  116.             for(j = 1; j < lWidth - 1; j++)
  117.             {
  118.                 if( *pmidtemp == 0)
  119.                 {
  120.                     pmid++;
  121.                     pmidtemp++;
  122.                     continue;
  123.                 }
  125.                 p3 = *(pmid + 1);
  126.                 p2 = *(pmid + 1 - lWidth);
  127.                 p1 = *(pmid - lWidth);
  128.                 p0 = *(pmid - lWidth -1);
  129.                 p7 = *(pmid - 1);
  130.                 p6 = *(pmid + lWidth - 1);
  131.                 p5 = *(pmid + lWidth);
  132.                 p4 = *(pmid + lWidth + 1);
  134.                 p1 *= 2;
  135.                 p2 *= 4;
  136.                 p3 *= 8;
  137.                 p4 *= 16;
  138.                 p5 *= 32;
  139.                 p6 *= 64;
  140.                 p7 *= 128;
  142.                 sum = p0 | p1 | p2 | p3 | p4 | p5 | p6 | p7;
  143.             //    sum = p0 + p1 + p2 + p3 + p4 + p5 + p6 + p7;
  144.                 if(deletemark[sum] == 1)
  145.                 {
  146.                     *pmid = 0;
  147.                     bStart = true;
  149.                     #ifdef SHOW
  150.                     cvSet2D(ColorSrc,i,j,cvScalar(0,0,0));
  152.                     cvNamedWindow("delcolor");
  153.                     cvShowImage("delcolor",ColorSrc);
  154.                     cvWaitKey(2);
  155.                     #endif
  156.                 }
  158.                 pmid++;
  159.                 pmidtemp++;
  160.             }
  162.             pmid++;
  163.             pmid++;
  164.             pmidtemp++;
  165.             pmidtemp++;
  166.         }
  168.         #ifdef SHOW
  169.         printf("过了一圈\n");
  170.         #endif
  171.     }
  173.     return true;
  174. }
  176. int main()
  177. {
  178.     //#define WRITERESULT
  179.     //#define WRITEIMAGE
  181.     IplImage* src = cvLoadImage("char2.png",0); //加载图像并灰度化
  182.     cvThreshold(src,src,100,255,CV_THRESH_BINARY);  //二值化图像  
  183.     unsigned char* imagedata ;
  185.     ColorSrc = cvLoadImage("char2.png",1);
  187.     cvNamedWindow("s");
  188.     cvShowImage("s" , src);
  191.     FILE* fp ;
  193.     #ifdef WRITEIMAGE
  194.         fp = fopen("data255.txt","rt+");
  195.     #endif
  197.     imagedata = (unsigned char*)malloc(sizeof(char)*src->width*src->height); //动态分配内存
  199.     int x , y;
  200.     for(y=0;y<src->height;y++)
  201.     {
  202.         unsigned char* ptr  = (unsigned char*)(src->imageData + y*src->widthStep);
  203.         for(x=0;x<src->width;x++)
  204.         {
  205.             imagedata[y*src->width+x] = ptr[x] > 0 ? 1 : 0;
  207.             #ifdef WRITEIMAGE
  208.                 if(ptr[x] > 0)
  209.                     fprintf(fp,"1");
  210.                 else
  211.                 {
  212.                     fprintf(fp,"0");
  213.                 }
  214.             #endif
  215.         }
  217.         #ifdef WRITEIMAGE
  218.             fprintf(fp,"\n");
  219.         #endif
  220.     }
  222.     #ifdef WRITEIMAGE
  223.         fclose(fp);
  224.     #endif
  226.     #ifdef WRITERESULT
  227.         fp = fopen("result.txt","rt+");
  228.     #endif
  230.     ThiningDIBSkeleton(imagedata,src->width,src->height);//进行细化操作
  232.     for(y=0;y<src->height;y++)
  233.     {
  234.         unsigned char* ptr  = (unsigned char*)(src->imageData + y*src->widthStep);
  235.         for(x=0;x<src->width;x++)
  236.         {
  237.             ptr[x] = imagedata[y*src->width + x]>0? 255 : 0;
  239.             #ifdef WRITERESULT
  240.                 if(ptr[x] > 0)
  241.                     fprintf(fp,"1");
  242.                 else
  243.                 {
  244.                     fprintf(fp,"0");
  245.                 }
  246.             #endif
  247.         }
  249.         #ifdef WRITERESULT
  250.             fprintf(fp,"\n");
  251.         #endif
  252.     }
  254.     #ifdef WRITERESULT
  255.         fclose(fp);
  256.     #endif
  258.     cvNamedWindow("src");
  259.     cvShowImage("src" , src);
  260.     cvWaitKey(0);
  262.     cvReleaseImage(&src);
  263.     cvReleaseImage(&ColorSrc);
  264.     free(imagedata);
  266.     return 0;
  267. }

7070 1 细化前后对比

博客参考https://www.cnblogs.com/zhazhiqiang/p/4487950.html

发表评论

表情：
评论列表 （有 0 条评论，536人围观）

还没有评论，来说两句吧...

相关阅读

    相关 opencv实现图像细化效果

    在图像处理中，有时候我们会想要提取图像的骨架，这是就需要对图像进行细化，opencv中没有直接进行细化的算法，网上大部分的细化算法都是基于以前IplImage结构的，对于想要使

    我不是女神ヾ我不是女神ヾ/ 0 赞/ 296 阅读

    相关 Opencv处理图像--细化

    Opencv处理图像--细化 细化的算法有很多种，但比较常用的算法是查表法 细化是从原来的图中去掉一些点，但仍要保持原来的形状。 实际上是保持原图的骨架。 代码如上

    秒速五厘米秒速五厘米/ 0 赞/ 537 阅读