Title :
Image Moment Based Registration Scheme Utilizing Support Vector Machine
Author_Institution :
Dept. of Autom., Shanghai Univ. of Eng. Sci., Shanghai, China
Abstract :
A novel image registration scheme is proposed in this paper. Six low order image moments are used as image global pattern features and are feed into support vector machine to estimate translation, rotation and scaling parameters. Simulation results show that the proposed registration scheme is accurate and robust to noise.
Keywords :
image registration; support vector machines; image global pattern features; image moment; image registration; support vector machine; Additive white noise; Discrete cosine transforms; Gaussian noise; Image registration; Neural networks; Noise robustness; Parameter estimation; Pattern recognition; Support vector machines; Testing; image moment; image registration; support vector machine;
Conference_Titel :
Information Assurance and Security, 2009. IAS '09. Fifth International Conference on
Conference_Location :
Xi´an
Print_ISBN :
978-0-7695-3744-3
DOI :
10.1109/IAS.2009.175
