WSEAS Transactions on Signal Processing
Print ISSN: 1790-5052, E-ISSN: 2224-3488
Volume 10, 2014
Image Classification Using Novel Set of Charlier Moment Invariants
Authors: , ,
Abstract: The use of the discrete orthogonal moments, as feature descriptors in image analysis and pattern recognition is limited by their high computational cost. To solve this problem, we propose, in this paper a new approach for fast computation of Charlier‟s discrete orthogonal moments. This approach is based on the use of recurrence relation with respect to variable x instead of order n in the computation of Charlier‟s discrete orthogonal polynomials and on the image block representation for binary images and intensity slice representation for gray-scale images. The acceleration of the computation time of Charlier moments is due to an innovative image representation, where the image is described by a number of homogenous rectangular blocks instead of individual pixels. A novel set of invariants moment based on the Charlier moments is also proposed. These invariants moment are derived algebraically from the geometric moment invariants and their computation is accelerated using image representation scheme. The proposed algorithms are tested in several well known computer vision datasets, regarding computational time, image reconstruction, invariability and classification. The performance of Charlier invariants moment used as pattern features for a pattern recognition and classification is compared with Hu and Legendre invariants moment.
Search Articles
Keywords: Charlier discrete orthogonal polynomials, Charlier moments, Charlier invariant moments, Image block representation, image slice representation, Fast computation, Image reconstruction, Pattern recognition, classification
Pages: 156-167
WSEAS Transactions on Signal Processing, ISSN / E-ISSN: 1790-5052 / 2224-3488, Volume 10, 2014, Art. #16