Image analysis using separable Krawtchouk-Tchebichef's moments

Abstract

In this paper, we will present a new method of image reconstruction using a new type of separable discrete orthogonal moments called the Krawtchouk-Tchebichef moments. The latter are based on the bivariate discrete orthogonal polynomials defined from the product of Krawtchouk and Tchebichef of discrete orthogonal polynomials with a variable. The new method of image reconstruction is made from the blocks of each slice of the image using the Krawtchouk-Tchebichef moments for small orders. By experiments we show the effectiveness of our method with respect to the global approach of image reconstruction and the possibility of reconstructing the image by the Krawtchouk-Tchebichef moments is compared to the classical moments of Tchebichef and Krawtchouk.