Title :
A relaxation evolutionary image restoration method
Author :
Wang, Zhe ; Yu, Yinglin
Author_Institution :
Res. Inst. of Radio & Autom., South China Univ. of Technol., Guangzhou, China
Abstract :
Evolution strategies (ESs) are global optimization models in the light of natural evolution. Using relaxation evolution strategy, a novel method for restoration of gray level images blurred by a known shift-invariant point spread function and corrupted by added Gaussian white noise is presented in this paper. With relaxation mutation and other improvements of traditional ESs, a practical evolution image restoration method is proposed. Computer simulation examples illustrate the usefulness of our method. Comparisons to other image restoration methods, such as inverse filter and neural restoration method, are also provided
Keywords :
Gaussian noise; genetic algorithms; image restoration; iterative methods; optical transfer function; relaxation theory; search problems; white noise; Gaussian white noise; global optimization models; gray level images; image restoration; iterative method; relaxation evolutionary; relaxation mutation; search space; shift-invariant point spread function; Automation; Computer simulation; Degradation; Electronic switching systems; Fault tolerance; Genetic mutations; Hopfield neural networks; Image converters; Image restoration; White noise;
Conference_Titel :
Systems, Man, and Cybernetics, 1997. Computational Cybernetics and Simulation., 1997 IEEE International Conference on
Conference_Location :
Orlando, FL
Print_ISBN :
0-7803-4053-1
DOI :
10.1109/ICSMC.1997.638096