Adpative Thresholding Function Image Denoising using the CSO Algorithm

Abstract

We propose herein a novel adaptive threshold function method for image denoising in the wavelet domain within the Competitive Swarm Optimizer (CSO) modeling of sub-band coefficients. In this approach, the stochastic global optimization techniques such as Competitive Swarm Optimizer (CSO)  Cuckoo Search (CS) algorithm, artificial bee colony (ABC),Genetic algorithm (GA), Jaya Algorithm and particle swarm optimization (PSO) technique. Also, there is the exploitation of different variants in order to learn about the parameters for adaptive thresholding functions that are required for optimizing the performances. The thresholding function is optimized to address salt and pepper, speckle, and Gaussian noise through the use of various evolutionary optimization algorithms to produce better de-noised images. It was found that the CSO algorithm algorithm-based denoising methods give better performance in terms of peak signal-to-noise ratio (PSNR) and Image Quality index  (IQI) as compared to Cuckoo-based de-noising approach, which is effective in de-noising. Comparative results of the peak signal-to-noise ratio and image quality index demonstrate the robustness of the proposed optimization algorithm.