Title: An approach for rigid image registration based on wavelet decomposition and modified particle swarm optimisation

Authors: Wang Hu; An-long Chen; Li-li Song

Addresses: School of Computer Science and Engineering, University of Electronic Science and Technology of China, No. 2006, Xianyuan Road, Chengdu City, 611731, China. ' School of Computer Science and Engineering, University of Electronic Science and Technology of China, No. 2006, Xianyuan Road, Chengdu City, 611731, China. ' School of Computer Science and Technology, SouthWest University of Science and Technology, No. 59, Qinglong Road, Mianyang City, Sichuan Province, 621010, China

Abstract: An intelligent approach based on modified particle swarm optimisation and wavelet decomposition was proposed for rigid image registration. In the phase of approximate registration, the modified particle swarm optimisation (PSO), more efficient optimisation without the velocity item in the evolutionary equation, was employed to optimise the solution responding to the mutual information value calculated from the approximate sub-images extracted from the original images by wavelet decomposition. And in the phase of fine registration, cyclic variable method was employed to search maximal mutual information value calculated from the original images. The experimental result showed that the average total registration time was reduced greatly by wavelet decomposition and precision was improved by the modified PSO.

Keywords: wavelet decomposition; particle swarm optimisation; modified PSO; mutual information; rigid image registration; cyclic variable method.

DOI: 10.1504/IJMIC.2011.043150

International Journal of Modelling, Identification and Control, 2011 Vol.14 No.4, pp.272 - 278

Published online: 21 Mar 2015 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article