Denoising method of strong noise pollution image on basis of partial differential equation

Abstract

The invention discloses a denoising method of a strong noise pollution image on the basis of partial differential equation, mainly solving the problem of poor traditional denoising effect of the strong noise pollution image. The realization process comprises: (1) pretreating an input noise image u0, and recording the result as u; (2) calculating the partial derivative sum of the image u; (3) calculating the gradient module value of the image u; (4) according to the gradient and the gradient module value, building the partial differential equation; (5) calculating diffusion coefficients and psi in the partial differential equation; (6) utilizing the coefficients and psi, solving the partial differential equation to obtain a filter image; (7) calculating the peak signal to noise ratio PSNR of the filter image; and (8) repeating steps 2 to 7, when the PSNR value of the filter image output iteratively one time is smaller than that output iteratively in the previous time, stopping iteration, and outputting the previously iterated filter image. The invention has simple calculation and high operation speed, can better keep image texture details while smoothening strong noise and can be used for denoising a natural image with strong noise pollution.

Description

technical field The invention belongs to the technical field of image processing, relates to an image denoising method, and is suitable for noise removal of SAR images and natural images. Background technique Image denoising aims to process the noise-contaminated image through an algorithm to reduce the impact of noise on the original useful information and restore an image closer to the idealized image as much as possible. The preprocessing technology often used in image processing in the fields of change research, environmental disaster assessment, urban planning, national defense military monitoring, medical imaging and astronomical imaging has urgent needs and broad application prospects. Synthetic aperture radar SAR images and natural images both need denoising processing, and research on SAR image and natural image denoising processing technology has a very broad application prospect. In order to meet the urgent needs of image denoising applications, a lot of denoisi...

AI Technical Summary

Problems solved by technology

Although these denoising methods can better achieve the purpose of denoising, but for images with very low image details, the denoising results of these methods are not ideal.

In order to solve the problems of the above methods, denoising methods based on partial differential equations have become a hot issue in the field of image denoising research. Many scholars have classified, analyzed and improved the existing denoising methods from different perspectives, but for image Different feature areas, the effect of denoising is still not very ideal

Especially for some images with relatively high noise, the denoising results of these existing methods are poor and cannot meet the requirements

Images