Comparative Analysis of Image Denoising Methods Based on Wavelet Transform and Threshold Functions

Abstract

There are many unavoidable noise interferences in image acquisition and transmission. To make it better for subsequent processing, the noise in the image should be removed in advance. There are many kinds of image noises, mainly including salt and pepper noise and Gaussian noise. This paper focuses on the research of the Gaussian noise removal. It introduces many wavelet threshold denoising algorithms which include global threshold denoising, Maxmin threshold denoising, and BayesShrink threshold denoising. We emphatically analyze the strengths and weaknesses of different denoising methods based on different threshold functions. Besides, we make a comparative analysis for these denoising methods. The experimental result shows that the wavelet images denoising algorithm based on Gaussian mixture model is better than that of the global threshold and Maxmin threshold, and also slightly better than BayesShrink threshold.