Iterative noise reduction method based on image low-rank performance

Abstract

The invention provides an iterative noise reduction method based on image low-rank performance. The method concretely relates to the field of image processing and computer vision. In the image representation, Y refers to an image having noise, X refers to an ideal image, N refers to Gaussian white noise of which the standard deviation is tau, and the image is described as Y=X+N and denoising means to reduce the value of N. In order to better utilize the image low-rank performance to perform denoising, firstly an observation image is divided into multiple small blocks which are identical in size. As for any image block P, image blocks similar to the image block P are searched according to a certain similarity criterion. The similar image blocks are vectorized so that a similar block matrix P is formed, based on which the P is estimated by utilizing the low-rank performance of the similar block matrix and the theory of minimum variance estimation so that the objective of suppressing image block noise can be achieved. Finally the denoised image is obtained through reconstruction of the image block estimation value. As for the problem of a small amount of residual noise, the denoised image is processed by adopting a synthetic iterative method.

Description

technical field [0001] The method relates to the field of computer information technology, in particular to the fields of image processing and computer vision. Background technique [0002] Affected by imaging conditions and external interference factors, different degrees of noise will be generated in the process of image acquisition, compression, and transmission, which reduces the visual quality of the image and affects the subsequent processing of the image. Therefore, image noise reduction has been used in image processing and Fields such as computer vision have received greater attention. [0003] In terms of research, some scholars use estimation methods and wavelet coefficients to propose a denoising method based on wavelet transform, and some scholars use generalized Gaussian distribution to model wavelet coefficients, and then use Bayesian estimation to derive image shrinkage coefficients for noise reduction. The bivariate denoising model is established by using t...

AI Technical Summary

Problems solved by technology

[0005] In view of the above shortcomings, the present invention proposes an iterative denoising method based on low-rank image, which utilizes image self-similarity to construct a similar block matrix with low-rank property, thereby transforming the image denoising problem into a low-rank matrix For the estimation problem, an effective image denoising method is derived based on the minimum variance estimation theory. At the same time, for a small amount of noise residual problem, an iterative method is used to further improve the denoising performance of the method

Images