Nuclear magnetic resonance image denoising method based on image-block self-similarity prior

Abstract

The invention discloses a nuclear magnetic resonance image denoising method based on image-block self-similarity prior. The method comprises the following steps of based on self-similarity among imageblocks, constructing a clustering regular term; based on the clustering regular term and a low rank decomposition denoising method, establishing an image denoising model; and optimizing the image denoising model, and based on the optimized image denoising model, carrying out denoising processing on an image. Disadvantages in an existing denoising method are overcome. Noises are removed and simultaneously detail information of the image can be kept to a greater degree.

Description

technical field [0001] The invention relates to the field of image denoising processing, in particular to a nuclear magnetic resonance image denoising method based on image block self-similarity prior. Background technique [0002] MRI images (Magnetic Resonance images, MRI) are an important medical tool to help doctors diagnose patients' conditions. Doctors can accurately and quickly confirm patients' conditions through MRI to ensure the best time for treatment. However, MRI is susceptible to noise pollution during the imaging process. These noises will greatly affect the quality of the image and seriously affect the accuracy of the doctor's diagnosis of the patient's condition. [0003] MRI image denoising is a typical ill-posed problem in image processing. For the ill-posed problem in image processing, the existing method is to add image prior information to make it well-conditioned. Common image denoising methods use properties such as image self-similarity, redundancy,...

AI Technical Summary

Problems solved by technology

[0005] (1) The method of denoising the image according to the image grayscale and image gradient information is an operation for the entire image, which makes the optimization of the entire image in the denoising process a huge workload and takes a long time;

[0006] (2) With the increase of noise intensity, the ability of most existing denoising methods to preserve image details is greatly reduced, and the denoising results do not conform to human visual perception

Images