Non-uniform noise image denoising method based on non-local mean

Abstract

The invention discloses a non-uniform noise image denoising method based on non-local mean value. The method comprises the following steps: 1. Pixel rough classification of the non-uniform noise imageis carried out by using an evaluation operator R; 2, classifying each pixel in the noise image into one of low noise high texture, medium texture, high noise sub-texture and smooth area by adopting amajority voting method according to the rough classification result of neighboring pixels around the pixel, wherein the pixel is divided into three types: low noise high texture, medium texture, highnoise sub-texture and smooth area; 3, adaptively selecting filter parameters and neighborhood block sizes for each category after subclassification, and denoising pixels by using a non-local mean denoising algorithm; Achieve a balance between noise cancellation and texture preservation.

Description

technical field [0001] The invention relates to the field of image processing, in particular to a non-uniform noise image denoising method based on non-local means. Background technique [0002] Restricted by the hardware conditions of the camera, the digital image captured by the digital camera has noise, and the noise content on each color channel of the digital image is uneven. Therefore, the captured color image is polluted by non-uniform noise. Most existing denoising methods focus on additive white Gaussian noise (AWGN), where the observed noisy image is modeled as a clean image and the addition of AWGN, i.e., z(i)=x(i)+n( i), assuming that the noise variance on the entire image is fixed, and then denoising the image. In doing so, there will inevitably be deviations in the follow-up experiment process, which will also have a certain impact on the follow-up research. In 2005, Buades et al. proposed a non-local means (NLM) denoising algorithm, the basic idea of ​​whic...

AI Technical Summary

Problems solved by technology

[0008] For the denoising problem of non-uniform noise images, the present invention provides a non-uniform noise image de-noising method based on non-local means, which solves the existing deficiencies, and uses an evaluation operator to simultaneously evaluate the texture intensity of the local area of ​​the image and noise content are described; according to the description value, the image pixels are first roughly classified into flat areas and texture areas, and then the voting strategy is used to fine-tune the image pixels, and finally heuristic denoising parameters are selected for each type of area , in order to achieve a balance between the effect of noise reduction and the effect of texture preservation

Images