Opencv gray histogram, histogram specification

This paper mainly introduces the processing of gray histogram correlation , It includes the following aspects ：

• utilize OpenCV Calculate the gray histogram of the image , And draw histogram curve
• Principle and implementation of histogram equalization
• Histogram specification （ matching ） Principle and implementation of

Gray histogram of the image

Implementation of histogram specification

The implementation of histogram specification can be divided into the following three steps ：
-  Calculate the cumulative histogram of the original image
-  Calculate the cumulative histogram of the specified histogram
-  Calculate the absolute value of the difference between the two cumulative histograms
-  The gray level mapping is established according to the cumulative histogram difference

void HistSpecify(const Mat& src, const Mat& ref, Mat& result)
	{
		Histogram1D hist1D;
		Mat src_hist = hist1D.getHistogram(src);
		Mat dst_hist = hist1D.getHistogram(ref);

		float src_cdf[256] = { 0 };
		float dst_cdf[256] = { 0 };

		//  Normalize the histogram 
		src_hist /= (src.rows * src.cols);
		dst_hist /= (ref.rows * ref.cols);

		//  Calculate the cumulative probability of the original histogram and the specified histogram 
		for (int i = 0; i < 256; i++)
		{
			if (i == 0)
			{
				src_cdf[i] = src_hist.at<float>(i);
				dst_cdf[i] = dst_hist.at<float>(i);
			}
			else
			{
				src_cdf[i] = src_cdf[i - 1] + src_hist.at<float>(i);
				dst_cdf[i] = dst_cdf[i - 1] + dst_hist.at<float>(i);
			}
		}

		//  The difference of cumulative probability 
		float diff_cdf[256][256];
		for (int i = 0; i < 256; i++)
			for (int j = 0; j < 256; j++)
				diff_cdf[i][j] = fabs(src_cdf[i] - dst_cdf[j]);

		//  Build a gray level mapping table 
		Mat lut(1, 256, CV_8U);
		for (int i = 0; i < 256; i++)
		{
			//  The gray level of the search source is ｉ Mapped grayscale 
			//　 and ｉ The specified gray scale with the minimum cumulative probability difference 
			float min = diff_cdf[i][0];
			int index = 0;
			for (int j = 1; j < 256; j++)
			{
				if (min > diff_cdf[i][j])
				{
					min = diff_cdf[i][j];
					index = j;
				}
			}
			lut.at<uchar>(i) = static_cast<uchar>(index);
		}

		//  Apply lookup table , Make histogram specification 
		LUT(src, lut, result);
	}

 

summary

  The gray histogram of the image can intuitively show the overall distribution of gray levels in the image , It has a good guiding role for the subsequent image processing .-  Histogram equalization is to flatten the histogram of an image , Make each gray level tend to be evenly distributed , This can enhance the image contrast very well . Histogram equalization is an automatic transformation , Just input the image , The transformation function of the image can be determined . But the equalization operation of histogram is also certain , In the process of equalization, the data in the image is not selected , This may enhance the background of the image ; The gray level of the transformed image is reduced , It may cause some details to disappear ; It will compress the peak in the image histogram , Cause the unnatural contrast of the processed image .-  Histogram specification , Also known as histogram matching , After specified processing, the histogram of the original image is transformed into a histogram of a specific shape （ Example in above , It is to transform the histogram of an image into the histogram of another image ）. It can adjust the histogram of the image according to a preset shape , Based on the principle of equalization , By establishing the relationship between the original image and the desired image , Selectively control the histogram , Make the histogram of the original image into a specified shape. It can adjust the histogram of the image according to a preset shape . Histogram specification is based on the principle of equalization , By establishing the relationship between the original image and the desired image , Selectively control the histogram , Make the histogram of the original image into a specified shape , So as to make up for some shortcomings of histogram equalization .

Basis of image processing (8)： Gray histogram of the image 、 Histogram equalization 、 Histogram specification （ matching ） - Brook_icv - Blog Garden