期刊名称：BVICAM's International Journal of Information Technology
印刷版ISSN：0973-5658
出版年度：2013
卷号：5
期号：2
语种：English
出版社：Bharati Vidyapeeth's Institute of Computer Applications and Management
摘要：Degradation of images due to noise has led to the formulation of various techniques for image restoration. Wavelet shrinkage image denoising being one such technique has been improved over the years by using Particle Swarm Optimization (PSO) and its variants for optimization of the wavelet parameters. However, the use of PSO has been rendered ineffective due to premature convergence and failure to maintain good population diversity. This paper proposes a Hénon map based adaptive PSO (HAPSO) for wavelet shrinkage image denoising. While significantly improving the population diversity of the particles, it also increases the convergence rate and thereby the precision of the denoising technique. The proposed PSO uses adaptive cognitive and social components and adaptive inertia weight factors. The Hénon map sequence is applied to the control parameters instead of random variables, which introduces ergodicity and stochastic property in the PSO. This results in a more improved global convergence as compared to the traditional PSO and classical thresholding techniques. Simulation results and comparisons with the standard approaches show the effectiveness of the proposed algorithm.
