Image Denoising with Adaptive Weighted Graph Filtering

作     者：Ying Chen Yibin Tang Lin Zhou Yan Zhou Jinxiu Zhu Li Zhao 

作者机构：School of Information Science and EngineeringSoutheast UniversityNanjing210096China Department of Psychiatry and Translational ImagingColumbia University&NYSPINew York10032USA College of Internet of Things EngineeringHohai UniversityChangzhou213022China Changzhou Key Laboratory of Sensor Networks and Environmental SensingChangzhou213022China 

出 版 物：《Computers, Materials & Continua》 (计算机、材料和连续体（英文）)

年 卷 期：2020年第64卷第8期

页      面：1219-1232页

核心收录：

学科分类：0831[工学-生物医学工程（可授工学、理学、医学学位）] 0808[工学-电气工程] 0809[工学-电子科学与技术（可授工学、理学学位）] 07[理学] 0805[工学-材料科学与工程（可授工学、理学学位）] 0701[理学-数学] 0801[工学-力学（可授工学、理学学位）] 0812[工学-计算机科学与技术（可授工学、理学学位）] 070101[理学-基础数学] 

基　　金：This work is supported by National Natural Science Foundation of China The initials of authors who received these grants are LZ and YZ,respectively.It is also supported by Natural Science Foundation of Jiangsu Province,China[BK20170306] The initials of author who received this grant are YZ 

主　　题：Graph filtering image denoising Laplacian matrix low rank 

摘      要：Graph filtering,which is founded on the theory of graph signal processing,is proved as a useful tool for image *** graph filtering methods focus on learning an ideal lowpass filter to remove noise,where clean images are restored from noisy ones by retaining the image components in low graph frequency ***,this lowpass filter has limited ability to separate the low-frequency noise from clean images such that it makes the denoising procedure less *** address this issue,we propose an adaptive weighted graph filtering(AWGF)method to replace the design of traditional ideal lowpass *** detail,we reassess the existing low-rank denoising method with adaptive regularizer learning(ARLLR)from the view of graph filtering.A shrinkage approach subsequently is presented on the graph frequency domain,where the components of noisy image are adaptively decreased in each band by calculating their component *** a result,it makes the proposed graph filtering more explainable and suitable for ***,we demonstrate a graph filter under the constraint of subspace representation is employed in the ARLLR ***,ARLLR can be treated as a special form of graph *** not only enriches the theory of graph filtering,but also builds a bridge from the low-rank methods to the graph filtering *** the experiments,we perform the AWGF method with a graph filter generated by the classical graph Laplacian *** results show our method can achieve a comparable denoising performance with several state-of-the-art denoising methods.