Optimization based Filtering of Random Valued Impulses

This book describes the concepts of image de noising methods with optimization of the operators through easy-to-understand steps and schematic diagrams. Introductory chapter describes the basics of image processing including various existing methods of image de noising both in spatial and frequency domains. An extensive literature survey along with analysis of the existing schemes and their visual restoration results for the benchmark images are described. Among two most popular noise models, Random Valued Impulse Noise model has been taken for restoration purpose in this book. Filtering methods based on directional weighted minimum variance, all neighbor directional weighted pixels, fixed and variable mask median are presented in this book. Optimizations of the methods are done using two recent stochastic global optimization techniques such as GA and PSO, along with analysis of simulation results and comparisons with recent operators. All of these features make this book ideal for image de noising and related applications in the field of medical, satellite and under water image processing. Practicing engineers would also find this book to be a valuable reference source.