基于伽马范数最小化的图像去噪算法

doi:10.11959/j.issn.1000-436x.2020190

Abstract

Abstract:

Focusing on the issue of rather poor denoising performance of the traditional kernel norm minimization based method caused by the biased approximation of kernel norm to rank function,based on the low-rank theory,a gamma norm minimization based image denoising algorithm was developed.The noisy image was firstly divided into some overlapping patches via the proposed algorithm,and then several non-local image patches most similar to the current image patch were sought adaptively based on the structural similarity index to form the similar image patch matrix.Subsequently,the non-convex gamma norm could be exploited to obtain unbiased approximation of the matrix rank function such that the low-rank denoising model could be constructed.Finally,the obtained low-rank denoising optimization issue could be tackled on the basis of singular value decomposition,and therefore the denoised image patches could be re-constructed as a denoised image.Simulation results demonstrate that,compared to the existing state-of-the-art PID,NLM,BM3D,NNM,WNNM,DnCNN and FFDNet algorithms,the developed method can eliminate Gaussian noise more considerably and retrieve the original image details rather precisely.

Key words: image denoising, low-rank denoising model, non-convex optimization, gamma norm, structural similarity index

CLC Number:

TP391.41

Hongyan WANG,Tuo WANG,Mian PAN,Zumin WANG. Gamma norm minimization based image denoising algorithm[J]. Journal on Communications, 2020, 41(10): 222-234.

Figures/Tables 7

References 30

[1]	YAO S K , CHANG Y , QIN X J ,et al. Principal component dictionary-based patch grouping for image denoising[J]. Journal of Visual Communication and Image Representation, 2018,50(2): 111-122.
[2]	YAIR N , MICHAELI T . Multi-scale weighted nuclear norm image restoration[C]// 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE Press, 2018: 3165-3174.
[3]	ZHA Z Y , ZHANG X G , WANG Q ,et al. Group sparsity residual constraint for image denoising with external nonlocal self-similarity prior[J]. Neurocomputing, 2018,275(2): 2294-2306.
[4]	LI H , TANG G J , LIU X H ,et al. An enhanced image denoising method using method noise[C]// 2017 IEEE International Conference on Signal Processing. Piscataway:IEEE Press, 2017: 1-6.
[5]	PORTILLA J , STRELA V , WAINWRIGHT M J ,et al. Image denoising using scale mixtures of Gaussians in the wavelet domain[J]. IEEE Transactions on Image processing, 2003,12(11): 1338-1351.
[6]	CHAMBOLLE A . An algorithm for total variation minimization and applications[J]. Journal of Mathematical Imaging and Vision, 2004,20(1-2): 89-97.
[7]	KNAUS C , ZWICKER M . Progressive image denoising[J]. IEEE Transactions on Image Processing, 2014,23(7): 3114-3125.
[8]	FOI A , BORACCHI G . Foveated nonlocal self-similarity[J]. International Journal of Computer Vision, 2016,120(1): 78-110.
[9]	BUADES A , COLL B , MOREL J M . A non-local algorithm for image denoising[C]// 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE Press, 2005: 60-65.
[10]	DABOV K , FOI A , KATKOVNIK V ,et al. Image denoising by sparse 3-D transform-domain collaborative filtering[J]. IEEE Transactions on Image Processing, 2007,16(8): 2080-2095.
[11]	DONG W S , ZHANG L , SHI G M ,et al. Nonlocally centralized sparse representation for image restoration[J]. IEEE Transactions on Image Processing, 2013,22(4): 1620-1630.
[12]	KUMAR A , AHMAD M O , SWAMY M N S . An efficient denoising framework using weighted overlapping group sparsity[J]. Information Sciences, 2018,454(2): 292-311.
[13]	WANG S L , ZHANG L , LIANG Y . Nonlocal spectral prior model for low-level vision[C]// Asian Conference on Computer Vision. New York:ACM Press, 2012: 231-244.
[14]	JI H , LIU C Q , SHEN Z W ,et al. Robust video denoising using low rank matrix completion[C]// 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE Press, 2010: 1791-1798.
[15]	GU S H , XIE Q , MENG D Y ,et al. Weighted nuclear norm minimization and its applications to low level vision[J]. International Journal of Computer Vision, 2017,121(2): 183-208.
[16]	CHO S I , KANG S J . Gradient prior-aided CNN denoiser with separable convolution-based optimization of feature dimension[J]. IEEE Transactions on Multimedia, 2018,21(2): 484-493.
[17]	ZHANG K , ZUO W M , CHEN Y J ,et al. Beyond a gaussian denoiser:residual learning of deep CNN for image denoising[J]. IEEE Transactions on Image Processing, 2017,26(7): 3142-3155.
[18]	ZHANG K , ZUO W M , ZHANG L . FFDNet:toward a fast and flexible solution for CNN-based image denoising[J]. IEEE Transactions on Image Processing, 2018,27(9): 4608-4622.
[19]	BINH P H T , CRUZ C , EGIAZARIAN K . Flashlight CNN image denoising[J]. arXiv preprint,arXiv:2003.00762, 2020
[20]	WANG S S , LIU D H , ZHANG Z H . Nonconvex relaxation approach es to robust matrix recovery[C]// International Joint Conference on Artificial Intelligence. Piscataway:IEEE Press, 2013: 1764-1770.
[21]	WANG H Y , CEN Y G , HE Z Q ,et al. Reweighted low-rank matrix analysis with structural smoothness for image denoising[J]. IEEE Transactions on Image Processing, 2018,27(4): 1777-1792.
[22]	STARK C J . Learning optimal quantum models is NP-hard[J]. Physical Review A, 2018,97(2): 97-103.
[23]	CAI J F,CANDèS E J , SHEN Z W . A singular value thresholding algorithm for matrix completion[J]. SIAM Journal on Optimization, 2010,20(4): 1956-1982.
[24]	ZHANG H M , YANG J , XIE J C ,et al. Weighted sparse coding regularized nonconvex matrix regression for robust face recognition[J]. Information Sciences, 2017,394(2): 1-17.
[25]	DONG W S , SHI G M , LI X . Nonlocal image restoration with bilateral variance estimation:a low-rank approach[J]. IEEE Transactions on Image Processing, 2013,22(2): 700-711.
[26]	WANG Z , BOVIK A C , SHEIKH H R ,et al. Image quality assessment:from error visibility to structural similarity[J]. IEEE Transactions on Image Processing, 2004,13(4): 600-612.
[27]	ZHANG C H . Nearly unbiased variable selection under minimax concave penalty[J]. The Annals of Statistics, 2010,38(2): 894-942.
[28]	OSHER S , BURGER M , GOLDFARB D ,et al. An iterative regularization method for total variation based image restoration[C]// 2011 IEEE International Conference on Imaging Systems and Techniques. Piscataway:IEEE Press, 2011: 170-175.
[29]	THU Q , GHANBARI M . Scope of validity of PSNR in image/video quality assessment[J]. Electronics letters, 2008,44(13): 800-801.
[30]	PALUBINSKAS G , . Mystery behind similarity measures MSE and SSIM[C]// 2014 IEEE International Conference on Image Processing (ICIP). Piscataway:IEEE Press, 2014: 575-579.

Metrics

Recommended 0

No Suggested Reading articles found!

去噪算法	σ_n	Monarch/dB	Boats/dB	Lena/dB	Peppers/dB	House/dB	Barbara/dB	平均值
	10	34.18	35.95	33.40	35.01	36.60	35.04	35.03
PID	30	28.63	30.68	29.58	31.39	32.06	30.12	30.41
	50	26.21	28.25	27.85	29.38	29.48	27.56	28.12
	100	22.59	25.08	25.37	25.99	25.32	24.04	24.73
	10	32.54	34.79	32.79	34.20	35.02	33.91	33.87
NLM	30	27.56	28.63	28.28	29.42	29.18	27.32	28.40
	50	24.32	25.61	25.66	26.31	25.76	24.13	25.30
	100	19.68	21.32	21.49	21.58	21.09	20.45	20.94
	10	34.12	36.17	33.58	34.81	36.55	35.60	35.14
BM3D	30	28.36	30.79	29.54	31.11	31.94	30.24	30.33
	50	25.82	28.39	27.75	28.98	29.51	27.56	28.00
	100	22.52	25.26	25.22	25.79	25.54	23.84	24.69
	10	33.54	33.05	35.19	33.77	35.97	34.40	34.32
NNM	30	27.44	27.82	30.15	28.11	30.99	28.59	28.85
	50	24.46	25.39	27.74	25.29	27.84	25.75	26.08
	100	20.22	22.48	24.41	21.24	23.65	22.14	22.36
	10	35.26	33.99	35.91	34.87	36.45	33.15	34.94
DnCNN	30	29.42	28.34	30.18	29.85	32.29	28.69	29.80
	50	26.85	26.00	27.75	27.35	29.91	26.61	27.41
	100	14.07	13.85	13.93	14.01	13.98	14.03	13.98
	10	35.07	34.01	36.01	34.79	36.53	33.78	35.03
FFDNet	30	29.39	28.37	30.36	29.87	32.57	28.71	29.88
	50	26.92	26.06	27.98	27.41	30.36	26.69	27.56
	100	23.40	23.27	25.01	24.03	26.98	23.70	24.40
	10	35.04	33.73	35.72	34.95	36.93	35.15	35.25
WNNM	30	28.91	27.88	29.92	29.45	32.52	29.22	29.65
	50	26.32	25.54	27.48	26.91	30.33	26.64	27.20
	100	22.95	22.74	24.53	23.45	26.66	23.31	23.94
	10	35.28	36.34	33.38	35.10	36.64	36.17	35.47
所提算法	30	28.89	30.84	29.57	31.51	32.67	30.86	30.72
	50	26.32	28.49	27.84	29.63	30.52	28.25	28.51
	100	22.99	25.46	25.62	26.08	26.55	24.51	25.20

Gamma norm minimization based image denoising algorithm

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 7

References 30

Related Articles 4

Metrics

Recommended 0

算法	平均时间/s
PID	202
NLM	65
BM3D	2
NNM	212
所提算法	216
DnCNN	0.65
FFDNet	0.4
WNNM	317

[1]	Hongyan WANG, Xiao YANG, Yanchao JIANG, Zumin WANG. Image denoising algorithm based on multi-channel GAN [J]. Journal on Communications, 2021, 42(3): 229-237.
[2]	Mei-ling WANG,Xian-chun ZHOU,Lin-feng ZHOU,Lan-fang SHI. Coupling image denoising model based on total variation [J]. Journal on Communications, 2016, 37(4): 182-191.
[3]	. ISI sparse channel estimation based on ISL0 algorithm [J]. Journal on Communications, 2014, 35(5): 17-133.
[4]	Ting LIU,Jie ZHOU,CHIJiu-he JU. ISI sparse channel estimation based on ISL0 algorithm [J]. Journal on Communications, 2014, 35(5): 124-133.