数字图像的直方图均衡化（C/C++源代码）

数字图像的直方图均衡化是常用的图像增强方法，因为均衡化是自动完成的，无需人工干预，而且常常得到比较满意的结果。下面的程序是利用OPENCV提供的函数，实现这个功能。需要OPENCV B4.0的支持，在VC6下编译通过。

//
// perform histgram equalization for single channel image
// AssureDigit Sample code
//

#include "cv.h"
#include "highgui.h"

#define HDIM    256    // bin of HIST, default = 256

int main( int argc, char** argv )
{
    IplImage *src = 0, *dst = 0;
    CvHistogram *hist = 0;
   
    int n = HDIM;    
    double nn[HDIM];
    uchar T[HDIM];
    CvMat *T_mat;
   
    int x;
    int sum = 0; // sum of pixels of the source image 图像中象素点的总和
    double val = 0;
   
    if( argc != 2 || (src=cvLoadImage(argv[1], 0)) == NULL)  // force to gray image
        return -1;
   
    cvNamedWindow( "source", 1 );
    cvNamedWindow( "result", 1 );
   
    // calculate histgram 计算直方图
    hist = cvCreateHist( 1, &n, CV_HIST_ARRAY, 0, 1 ); 
    cvCalcHist( &src, hist, 0, 0 );
   
    // Create Accumulative Distribute Function of histgram
    val = 0;
    for ( x = 0; x < n; x++)
    {
        val = val + cvGetReal1D (hist->bins, x);
        nn[x] = val;
    }

    // Compute intensity transformation 计算变换函数的离散形式
    sum = src->height * src->width;
    for( x = 0; x < n; x++ )
    {
        T[x] = (uchar) (255 * nn[x] / sum); // range is [0,255]
    }

    // Do intensity transform for source image
    dst = cvCloneImage( src );
    T_mat = cvCreateMatHeader( 1, 256, CV_8UC1 );
    cvSetData( T_mat, T, 0 );   
    // directly use look-up-table function 直接调用内部函数完成 look-up-table 的过程
    cvLUT( src, dst, T_mat );

    cvShowImage( "source", src );
    cvShowImage( "result", dst );
    cvWaitKey(0);

    cvDestroyWindow("source");
    cvDestroyWindow("result");
    cvReleaseImage( &src );
    cvReleaseImage( &dst );
    cvReleaseHist ( &hist );
   
    return 0;
}