Title :
Deformed shape retrieval based on Markov model
Author :
Chang, Fa-Shyang ; Chen, Shu-Yuan
Author_Institution :
Dept. of Comput. Eng. & Sci., Yuan-Ze Univ., Chungli, Taiwan
fDate :
1/20/2000 12:00:00 AM
Abstract :
A new deformed shape retrieval method based on the Markov model (MM) is proposed. The proposed method is translation rotation and scale invariant. In addition, it is robust to various non-rigid deformations such as perspective, shear, occlusion, pincushion, barrel and other nonlinear distortions. The proposed method has been applied to two databases: geometric and character. The effectiveness and practicability of the proposed approach have been proven by various experimental results
Keywords :
Markov processes; character recognition; feature extraction; image recognition; image retrieval; visual databases; Markov model; character databases; deformed shape retrieval method; geometric databases;
Journal_Title :
Electronics Letters
DOI :
10.1049/el:20000149
