集美大学学报（自然科学版）2023，Vol.28Issue(5)：454-460,7.DOI:10.19715/j.jmuzr.2023.05.09
基于整体变分模型的随机值脉冲噪声去除方法
Removal Method of Random-Valued Impulse Noise Based on Total Variation Model
胡樾 1陶胜2
作者信息
- 1. 集美大学诚毅学院,福建 厦门 361021
- 2. 集美大学理学院,福建 厦门 361021
- 折叠
摘要
Abstract
To effectively detect random-valued impulse noise and remove its interference on the image,this paper proposes a new switching total variation denoising method.The method detects and modifies noise in different window neighborhoods of 3 px×3 px,5 px×5 px and 7 px×7 px according to different densities of random-valued impulse noise,and adopts a total variation based method to denoise the pixels that are finally judged as noise.In addition,the method uses a dynamic threshold to perform multiple noise detection and re-moval on the denoised image,thereby further improving the denoising effect.Simulation results show that the method can effectively remove noise and protect image details for images with random-valued impulse noise of different densities.It outperforms other methods in terms of peak signal noise ratio.关键词
随机值脉冲噪声/噪声检测/整体变分模型/图像去噪/峰值信噪比Key words
random-valued impulse noise/noise detection/total variation modal/image denoising/peak sig-nal noise ratio(PSNR)分类
信息技术与安全科学引用本文复制引用
胡樾,陶胜..基于整体变分模型的随机值脉冲噪声去除方法[J].集美大学学报（自然科学版）,2023,28(5):454-460,7.
