Radial Charlier moment invariants for 2D object/image recognition

Abstract

Radial Charlier moments as discrete orthogonal moments in the polar coordinate are better descriptor in image processing applications and pattern recognition. However, the translation and scale invariant property of these moments have not been studied due to its complexity of the problem. In this paper, we present a method to construct a set of rotation invariants extracted from radial Charlier moments, named radial Charlier moment invariants (RCMI). Experimental results show the efficiency and the robustness to reconstruction error (MSE), peak signal to noise ratio (PSNR) of the proposed method.