A Weighted Fidelity and Regularization-Based Method for Mixed or Unknown Noise Removal From Images on Graphs

被引:40
作者
Wang, Cong [1 ]
Yan, ZiYue [2 ]
Pedrycz, Witold [1 ,3 ,4 ]
Zhou, MengChu [5 ,6 ]
Li, ZhiWu [1 ,5 ]
机构
[1] Xidian Univ, Sch Electromech Engn, Xian 710071, Peoples R China
[2] Univ Southern Calif, Dept Comp Sci, Viterbi Sch Engn, Los Angeles, CA 90007 USA
[3] Univ Alberta, Dept Elect & Comp Engn, Edmonton, AB T6R 2V4, Canada
[4] King Abdulaziz Univ, Fac Engn, Jeddah 21589, Saudi Arabia
[5] Macau Univ Sci & Technol, Inst Syst Engn, Macau 999078, Peoples R China
[6] New Jersey Inst Technol, Helen & John C Hartmann Dept Elect & Comp Engn, Newark, NJ 07102 USA
基金
中国国家自然科学基金;
关键词
Tight wavelet frame; variational model; mixed or unknown noise; image on graph; image denoising; IMPULSE NOISE; LAPLACIAN REGULARIZATION; RESTORATION; SPARSE; TRANSFORM; POISSON; SURFACES; SYSTEMS;
D O I
10.1109/TIP.2020.2969076
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Image denoising technologies in a Euclidean domain have achieved good results and are becoming mature. However, in recent years, many real-world applications encountered in computer vision and geometric modeling involve image data defined in irregular domains modeled by huge graphs, which results in the problem on how to solve image denoising problems defined on graphs. In this paper, we propose a novel model for removing mixed or unknown noise in images on graphs. The objective is to minimize the sum of a weighted fidelity term and a sparse regularization term that additionally utilizes wavelet frame transform on graphs to retain feature details of images defined on graphs. Specifically, the weighted fidelity term with -norm is designed based on a analysis of the distribution of mixed noise. The augmented Lagrangian and accelerated proximal gradient methods are employed to achieve the optimal solution to the problem. Finally, some supporting numerical results and comparative analyses with other denoising algorithms are provided. It is noted that we investigate image denoising with unknown noise or a wide range of mixed noise, especially the mixture of Poisson, Gaussian, and impulse noise. Experimental results reported for synthetic and real images on graphs demonstrate that the proposed method is effective and efficient, and exhibits better performance for the removal of mixed or unknown noise in images on graphs than other denoising algorithms in the literature. The method can effectively remove mixed or unknown noise and retain feature details of images on graphs. It delivers a new avenue for denoising images in irregular domains.
引用
收藏
页码:5229 / 5243
页数:15
相关论文
共 61 条
[1]  
[Anonymous], 2018, J ENG SCI TECHNOL RE
[2]  
[Anonymous], SERIES APPL MATH
[3]   Segmentation and Restoration of Images on Surfaces by Parametric Active Contours with Topology Changes [J].
Benninghoff, Heike ;
Garcke, Harald .
JOURNAL OF MATHEMATICAL IMAGING AND VISION, 2016, 55 (01) :105-124
[4]   IMAGE RESTORATION: TOTAL VARIATION, WAVELET FRAMES, AND BEYOND [J].
Cai, Jian-Feng ;
Dong, Bin ;
Osher, Stanley ;
Shen, Zuowei .
JOURNAL OF THE AMERICAN MATHEMATICAL SOCIETY, 2012, 25 (04) :1033-1089
[5]   Error-Optimized Sparse Representation for Single Image Rain Removal [J].
Chen, Bo-Hao ;
Huang, Shih-Chia ;
Kuo, Sy-Yen .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2017, 64 (08) :6573-6581
[6]   Overlapping Community Change-Point Detection in an Evolving Network [J].
Cheng, Jiujun ;
Chen, Minjun ;
Zhou, MengChu ;
Gao, Shangce ;
Liu, Chunmei ;
Liu, Cong .
IEEE TRANSACTIONS ON BIG DATA, 2020, 6 (01) :189-200
[7]   A Reconfigurable and Scalable FPGA Architecture for Bilateral Filtering [J].
Dabhade, Swapnil Deelip ;
Rathna, G. N. ;
Chaudhury, Kunal Narayan .
IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2018, 65 (02) :1459-1469
[8]   Image denoising by sparse 3-D transform-domain collaborative filtering [J].
Dabov, Kostadin ;
Foi, Alessandro ;
Katkovnik, Vladimir ;
Egiazarian, Karen .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2007, 16 (08) :2080-2095
[9]   A Patch-Based Approach for Removing Impulse or Mixed Gaussian-Impulse Noise [J].
Delon, Julie ;
Desolneux, Agnes .
SIAM JOURNAL ON IMAGING SCIENCES, 2013, 6 (02) :1140-1174
[10]   The contourlet transform: An efficient directional multiresolution image representation [J].
Do, MN ;
Vetterli, M .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2005, 14 (12) :2091-2106