一种去除椒盐噪声的混合正则化模型  

A Mixed Regularization Model for Removing Salt-and-Pepper Noise

作　　者：王佳佳 唐利明 WANG Jiajia;TANG Liming

机构地区：[1]湖北民族大学,湖北恩施445000

出　　处：《科技创新与应用》2022年第26期1-7,共7页Technology Innovation and Application

基　　金：国家自然科学基金资助项目(62061016;61561019)。

摘　　要：全变分图像去噪模型能有效地复原图像的边缘,但是在平滑区域存在阶梯效应,且不能有效复原图像纹理。为解决上述问题,该文结合全变分正则化子和分数阶正则化子,提出一种去除椒盐噪声的混合正则化模型,模型复原后的图像有效地保留图像的边缘和纹理细节信息,抑制阶梯效应。该文采用交替方向乘子(ADMM)法进行求解,实验结果验证模型和算法的有效性。The full variational image denoising model can effectively restore the edge of the image,but there is a ladder effect in the smooth area,and can not effectively restore the image texture.In order to solve the above problems,combined with total variational regularizer and fractional regularizer,a mixed regularization model for removing salt and pepper noise is proposed in this paper.The image restored by this model effectively retains the image edge and texture detail information and suppresses the ladder effect.In this paper,the alternating direction multiplier method(ADMM) is used to solve the problem,and the experimental results verify the effectiveness of the model and algorithm.

关 键 词：分数阶变分模型 全变分模型 ADMM 椒盐噪声 图像去噪 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]