Automated Processing for Color Image Arrangement Based on Histogram Matching Using Gaussian Distribution

This paper describes a method for color image arrangement using Histogram Matching based on a parameter estimated Gaussian distribution (HMGD). In our previous papers, we have presented some methods for the estimation of parameters in Gaussian distribution from the curvature computation of the original cumulative histogram. However, since the histogram of original image is not always ideally shaped for the Gaussian distribution approximation, the parameters estimation method based on curvature computation may not work well. In this paper, we propose an improved parameter (especially variance) estimation method using regression analysis. This method firstly detects the part of peak in original image histogram by using curvature computation, and secondly carries out the regression analysis based on approximated curvature formula. This paper also shows the experimental results.