Image denoising method and image denoising system

Abstract

The invention discloses an image denoising method. The method includes the steps of initializing parameters and image inputs, performing discrete Fourier transformation on initialized images, solving a fractional diffusion-wave equation model of a discrete space based on iterative computations, performing discrete Fourier transformation on iterative results to obtain a final result. By the adoption of the method, real edge structures and important texture details of the images can be well maintained, and the method is superior to similar methods evaluated both from objective performance indexes and subjective visual perception.

Description

technical field [0001] The invention belongs to the technical field of image denoising, and in particular relates to an image denoising method and a denoising system thereof. Background technique [0002] The purpose of image denoising is to remove the noise of the observed image and obtain an image as close as possible to the original noise-free image. In the process of image acquisition or transmission, it is usually affected by the performance of the hardware itself or the environment such as light and temperature, which greatly reduces the quality of the image. In addition to affecting the visual experience, this also affects the subsequent higher-level image analysis, such as recognition and classification. Therefore, image denoising is almost an essential preprocessing operation before any image processing. [0003] Existing image denoising methods include spatial domain filtering (such as mean filtering, median filtering, bilateral filtering, non-local mean filterin...

AI Technical Summary

Problems solved by technology

[0004] First, the parabolic diffusion equation is used to achieve image denoising, but the diffusion characteristics are not conducive to maintaining the edge structure;

[0005] Second, the visual effect of the denoising result is not good. For example, the second-order diffusion equation model represented by the PM nonlinear diffusion model usually leads to a "ladder effect" in the resulting image, that is, areas with block constant brightness values, while the fourth-order diffusion Although the equation model can remove the "ladder effect", black and white bright spots will appear, which is called "spot effect".

Images