Local Feature Extraction Property of Krawtchouk Moment

Abstract

Abstract-Moment descriptors have long been applied in object recognition since the early years of the development of the moment theories. In recent years, the emergence of discrete orthogonal moments has substantially enriched the moment methods. In this research, we have focused our study on the local feature extraction of Krawtchouk moment and the relationship between the Region of Interest (ROI) of an image and the parameter, p, of the Krawtchouk polynomials. It has been observed that by changing parameter p of the Krawtchouk moment, the pair of reconstructed images from different orders of Krawtchouk moments with the complimentary parameters, p and 1-p, are centrosymmetric.

Real-time Computation of Krawtchouk Moments on Gray Images Using Block Representation

“…The Hahn moments are considered as a generalization of Tchebichef and Krawtchouk moments. TheTchebichef moments are able to capture the global features of an image, while the Krawtchouk moments capture local features [25]. The discriminative power of the moments has as a result the reconstruction of the image from a finite number of moment values [22,23].…”