Title :
Translation invariants of modified Chebyshev moments
Author :
Fu, Bo ; Zhou, Jian-zhong ; Chen, Wen-Qing ; Zhang, Guo-Jun
Author_Institution :
Coll. of Hydropower & Inf. Eng., Huazhong Univ. of Sci. & Technol., Wuhan, China
Abstract :
The translation invariants of moments based on polynomials can be achieved by many methods, such as geometric moments, Legendre moments, and so on. In this paper, based on the modified discrete Chebyshev polynomials, the modified discrete Chebyshev moments, which can remain invariant for translated images and are superior to classical Chebyshev moments in the aspect of representation capability, is proposed. And the performance of the proposed descriptors is experimentally verified by using a set of binary English, Chinese characters as well.
Keywords :
Chebyshev approximation; character recognition; image representation; polynomials; Chebyshev moments; binary Chinese characters; binary English characters; discrete orthogonal systems; image feature representation; image representation; image translation; modified discrete Chebyshev polynomials; polynomial moments; translation invariants; Chebyshev approximation; Computational complexity; Educational institutions; Equations; Hydroelectric power generation; Image analysis; Image reconstruction; Kernel; Pattern recognition; Polynomials; Discrete orthogonal systems; image feature representation; modified Chebyshev polynomials; translation invariants;
Conference_Titel :
Machine Learning and Cybernetics, 2005. Proceedings of 2005 International Conference on
Conference_Location :
Guangzhou, China
Print_ISBN :
0-7803-9091-1
DOI :
10.1109/ICMLC.2005.1527731
