Image denoising method based on weighted variational model

Abstract

The invention discloses an image denoising method based on a weighted variational model, and belongs to the technical field of graphic image processing. According to the method, the influence of Gaussian noise on the aspects of image gradient, frequency and the like is fully analyzed, and the adaptive weight is proposed for the first time and introduced into the designed WTV model, so that duringdenoising, high-frequency information such as edges is enhanced while low-frequency information is kept, and the structural similarity of the images is well kept. Besides, the design of the weight only depends on the input noise image and does not involve other parameters so that application and implementation are greatly facilitated. Compared with the prior art, the method has the advantages thata self-adaptive weight function is adopted, extra parameters are not introduced, the weights of different pixels are different, and the parameters do not need to be adjusted. The weight function canmaintain low-frequency information and enhance high-frequency information at the same time, the image structure can be better maintained, and the denoising effect is improved. Through experimental tests, the method has obvious advantages in the aspects of evaluation indexes PSNR and SSIM and operation time.

Description

technical field [0001] The invention relates to an image denoising method, in particular to an image denoising method based on a weighted variational model, and belongs to the technical field of image processing. Background technique [0002] With the development of the information age, a large amount of image information emerges, which brings a lot of convenience to people's work and life. At the same time, valuable information is easily polluted by noise, such as Gaussian noise, salt and pepper noise, speckle noise, and mixed noise, which hinder vision and effectiveness. Therefore, how to deal with the noise mixed in the image is one of the very important issues. [0003] As an effective method to restore image information, image denoising technology emerges at the historic moment. Image denoising methods can be divided into spatial domain methods and transform domain methods. Among them, the spatial domain method is based on the independence of noise to directly operat...

AI Technical Summary

Problems solved by technology

However, these methods still have shortcomings

The method based on the HOTV model, the introduction of middle and high order gradients, l p -al q The regular term of the more complex model requires a higher calculation cost. The two parameters introduced by the ATV model when defining the weight need to be adjusted, which increases the difficulty of numerical experiments.

Images