Histogram Analysis Method Oriented for Image Arrangement

This paper proposes a histogram shape detection method for image arrangement processing by histogram matching, using curvature computation based on Gaussian distribution. The curvature computation is performed over the cumulative histogram of input image. By this computation, we can not only detect the peaks and valleys in the histogram of input image, but also obtain the accompanying information on the peaks and valleys such as the sizes of height, width, depth and so on. In this paper, first, we present the information extraction method from image histogram using the curvature computation, based on the assumption that the peaks and valleys are approximated by the same number of Gauss density functions. Second, we show the experimental results and the effectiveness of the proposed method.