Re: Problem equalizing the histogram of an image

Hi visnavki,

There is a Histogram equalisation function in opencv, which you can
use if you wish. 

It is:

------------
EqualizeHist
------------

Equalizes histogram of grayscale image

void  cvEqualizeHist( const CvArr* src, CvArr* dst );

src
    The input 8-bit single-channel image. 
dst
    The output image of the same size and the same data type as src. 

The function cvEqualizeHist equalizes histogram of the input image
using the following algorithm:

1. calculate histogram H for src.
2. normalize histogram, so that the sum of histogram bins is 255.
3. compute integral of the histogram:
   H'(i) = sum0≤j≤iH(j)
4. transform the image using H' as a look-up table: dst(x,y)=H'(src(x,y))

The algorithm normalizes brightness and increases contrast of the image.


For example, consider this code:
=======================================================================

    // Load the color image into a temporary file
    IplImage* temp = cvLoadImage( "flower.jpeg", -1 );

    // src is the grayscale image, and dst is the equalized image
    IplImage *src = cvCreateImage( cvGetSize( temp ), IPL_DEPTH_8U, 1 );
    IplImage *dst = cvCreateImage( cvGetSize( temp ), IPL_DEPTH_8U, 1 );

    // Convert the image into a grayscale image
    cvCvtColor( temp, src, CV_BGR2GRAY );

    // Release the temp, which is not necessary hereafter
    cvReleaseImage( &temp );

    // Histogram Equalization
    cvEqualizeHist( src, dst );

    // Show the GrayScale image
    cvNamedWindow( "Grayscale", 1);
    cvShowImage( "Grayscale", src );

    // Show the Histogram Equalized image
    cvNamedWindow( "Equalized", 1 );
    cvShowImage( "Equalized", dst );
=======================================================================

Regards, :D
Paramesh.

--- In OpenCV@xxxxxxxxxxxxxxx, "visnavki" <visnavki@xxxx> wrote:
>
> Hey folks,
> 
> I hope someone of you can help me regarding my concern. The code I'm
> using for implementing an histogram equalization is shown below. But
> it does not work in any sense. Curious to me is, that even the
> max_value variable does not contain a reasonable value, so something
> must go wrong when calculating the histogram itself. But I am unable
> to locate the point of failure.
> 
> Here's the code:
> **********************************
>       IplImage* iplGrayScaleImg = cvCreateImage(cvGetSize(input_image),
> IPL_DEPTH_8U, 1);
>       IplImage* iplGrayScaleImgOrg = cvCloneImage(iplGrayScaleImg);
> 
>       int iNumOfDims = 1;
>       int iDimSizes = 256;
>       float fRange_0[] = {0,256};
>       float* fRanges[] = {fRange_0};
>       CvHistogram* cvhImgHist = cvCreateHist(iNumOfDims, &iDimSizes,
> CV_HIST_ARRAY, fRanges, 1);
>       cvCalcHist(&iplGrayScaleImg, cvhImgHist, 0, 0);
>       
>       uchar ucLutData[256];
>       CvMat* cvmLut = cvCreateMatHeader(1, 256, CV_8UC1);
>       
>       float max_value=0.0;
>       cvGetMinMaxHistValue(cvhImgHist, 0, &max_value, 0, 0);
> 
>       int i,j;
>       int iQ = 256;
>       long iSum;
>       for(i=0; i<iDimSizes; i++)
>       {
>               iSum = 0;
>               for(j=0; j<=i; j++)
>               {
>                       iSum += (long)(cvQueryHistValue_1D(cvhImgHist, j));
>               }
>               ucLutData[i] =
> (uchar)((iQ-1)*iSum/(iplGrayScaleImg->width*iplGrayScaleImg->height));
>       }
> 
>       cvSetData(cvmLut, &ucLutData, 0);
> 
>       cvLUT(iplGrayScaleImgOrg, iplGrayScaleImg, cvmLut);
> 
>       output_image = cvCloneImage(iplGrayScaleImg);
> **********************************
> 
> Thanks for help in advance,
> 
> Chris
>







Change settings: http://www.yahoogroups.com/mygroups, select
   Get Emails (get all posts)
   Daily Digest (one summary email per day)
   Read on the web (read posts on the web only)Or Unsubscribe by mailing 
OpenCV-unsubscribe@xxxxxxxxxxxxxxx