An Image Denoising Method Based on Wavelet Spatial Correlation Edge Detection and Zerotree-Like Structure

Image denoising has been being a hotspot in the discipline of image processing. In this paper, we first analyze the intra-scale distribution and inter-scale distribution characteristics of the coefficients at finer-scale of the noise image, and then propose an edge detection method for preserving important high frequency information through incorporating the coefficients correlation of intra-scale and inter-scale dependencies. Second, we propose a novel image denoising algorithm by combining a zerotree-like structural Bayesian threshold denoising method with a wavelet spatial correlation edge detection method. Experimental results show that the proposed algorithm has steady and efficient result. Not only reserves more edge information, but also do better denoising performance than the traditional Bayes shrinkage method and zerotree-like structural Bayes threshold denoising method´s result.