关于二值化函数cvAdaptiveThreshold和cvThreshold的一些发现

// 自适应灰度二值化
//

#include "stdafx.h"
#include"highgui.h"
#include
#include
#include
#include

using namespace std;
using namespace cv;

int _tmain(int argc, _TCHAR* argv[])
{
 IplImage *src_image = 0;
   IplImage *dst_image = 0;
   IplImage *dst_image_adaptive = 0;
   int c;
   int threshold_type;
   int thresh_val;
   int max_val;
   int adaptive_method;
   int block_size;
   int offset;
  
   if( (src_image = cvLoadImage( "D:\\pictures\\lena_bw.bmp", 0)) == 0 )
      return -1;
  
   dst_image = cvCreateImage(cvSize(src_image->width,src_image->height), IPL_DEPTH_8U, 1);
   cvNamedWindow( "Threshold", CV_WINDOW_AUTOSIZE );

   dst_image_adaptive = cvCreateImage(cvSize(src_image->width,src_image->height), IPL_DEPTH_8U, 1);
   cvNamedWindow( "ThresholdAdaptive", CV_WINDOW_AUTOSIZE );

   threshold_type = CV_THRESH_BINARY;
   int threshold_typeBef = CV_THRESH_BINARY+1;
   cvCreateTrackbar( "threshold type", "Threshold", &threshold_type, 1, 0 );

   thresh_val = 100;
   int thresh_valBef = 0;
   cvCreateTrackbar( "thresh val", "Threshold", &thresh_val, 255, 0 );

   max_val = 100;
   int max_valBef = 0;
   cvCreateTrackbar( "max val", "Threshold", &max_val, 255, 0 );

   adaptive_method = CV_ADAPTIVE_THRESH_MEAN_C;
   int adaptive_methodBef = 1;
   cvCreateTrackbar( "adaptive method", "ThresholdAdaptive", &adaptive_method, 1, 0 );
     
   block_size = 1;
   int block_sizeBef = 0;
   cvCreateTrackbar( "block size", "ThresholdAdaptive", &block_size, 60, 0 );
        
   offset = 30;
   int offsetBef = 0;
   cvCreateTrackbar( "offset", "ThresholdAdaptive", &offset, 60, 0 );
           

   for(;;)
   {
      if((threshold_typeBef != threshold_type)    ||
         (thresh_valBef != thresh_val )          ||
         (max_valBef != max_val)                  ||
         (adaptive_methodBef != adaptive_method ) ||
         (block_sizeBef != block_size)            ||
         (offsetBef != offset))
      {
         cvAdaptiveThreshold(src_image, dst_image_adaptive, (double)max_val, adaptive_method,
                        threshold_type, block_size*2+3, (double)(offset-30));
        
         cvShowImage("ThresholdAdaptive", dst_image_adaptive);

         if(max_val < thresh_val )
         {
            max_val = thresh_val;
         }
         cvThreshold(src_image, dst_image, (double)thresh_val,
                  (double)max_val, threshold_type);
         cvShowImage("Threshold", dst_image);

        
         threshold_typeBef = threshold_type;
         thresh_valBef = thresh_val;
         max_valBef = max_val;
         adaptive_methodBef = adaptive_method;
         block_sizeBef = block_size;
         offsetBef = offset;

           
         //printf("The edge detection method is \n");
      }
      c = cvWaitKey(30);
      if( c == 'q' || c == 'Q' || (c & 255) == 27 || (c & 255) == 32 )
            break;
   }
  
   cvDestroyWindow( "Threshold" );//销毁窗口
   cvDestroyWindow( "ThresholdAdaptive" );//销毁窗口
   cvReleaseImage( &src_image ); //释放图像
   cvReleaseImage( &dst_image ); //释放图像
   cvReleaseImage( &dst_image_adaptive ); //释放图像
   return 0;

}

转载出处：