Poisson Noise Removal Based on Nonlocal Total Variation with Euler’s Elastica Pre-processing

doi:10.1007/s12204-017-1878-5

Journal of shanghai Jiaotong University (Science) ›› 2017, Vol. 22 ›› Issue (5): 609-614.doi: 10.1007/s12204-017-1878-5

Previous Articles Next Articles

Poisson Noise Removal Based on Nonlocal Total Variation with Euler’s Elastica Pre-processing

LIU Hongyi^a* (刘红毅), ZHANG Zhengrong^a (张峥嵘), XIAO Liang^b (肖亮), WEI Zhihui^b (韦志辉)

(a. School of Science; b. School of Computer Science and Technology, Nanjing University of Science and Technology, Nanjing 210094, China)
(a. School of Science; b. School of Computer Science and Technology, Nanjing University of Science and Technology, Nanjing 210094, China)

Online:2017-09-30 Published:2017-09-30
Contact: LIU Hongyi (刘红毅) E-mail: hyliu@njust.edu.cn

Abstract

Abstract: An enhancement-based Poisson denoising method for photon-limited images is presented. The noisy image is firstly pre-processed for enhancing incomplete object information, and then it is denoised while preserving the restored structural details. A variational regularization model based on Euler’s elastica (EE) is proposed for image enhancement pre-processing. A nonlocal total variation (NLTV) regularization model is then employed in the second stage of image denoising. The above two optimization problems are solved by the alternating direction method of multipliers (ADMM). For Poissonian images with low image peak values, experiments demonstrate the validity and efficiency of the proposed method for both restoring geometric structure and removing noise.

Key words: Poisson noise| nonlocal total variation (NLTV)| Euler’s elastica (EE)| variation regularization

摘要： An enhancement-based Poisson denoising method for photon-limited images is presented. The noisy image is firstly pre-processed for enhancing incomplete object information, and then it is denoised while preserving the restored structural details. A variational regularization model based on Euler’s elastica (EE) is proposed for image enhancement pre-processing. A nonlocal total variation (NLTV) regularization model is then employed in the second stage of image denoising. The above two optimization problems are solved by the alternating direction method of multipliers (ADMM). For Poissonian images with low image peak values, experiments demonstrate the validity and efficiency of the proposed method for both restoring geometric structure and removing noise.

关键词: Poisson noise| nonlocal total variation (NLTV)| Euler’s elastica (EE)| variation regularization

CLC Number:

TP 391

LIU Hongyia* (刘红毅), ZHANG Zhengronga (张峥嵘), XIAO Liangb (肖亮), WEI Zhihuib (韦志辉). Poisson Noise Removal Based on Nonlocal Total Variation with Euler’s Elastica Pre-processing[J]. Journal of shanghai Jiaotong University (Science), 2017, 22(5): 609-614.

References 17

[1]	FRYZLEWICZ P, NASON G P. A Haar-Fisz algorithmfor poisson intensity estimation [J]. Journal ofComputational & Graphical Statistics, 2004, 13(3):621-638.
[2]	DEY N, BLANC-FERAUD L, ZIMMER C, et al.Richardson-Lucy algorithm with total variation regularizationfor 3D confocal microscope deconvolution[J]. Microscopy Research & Technique, 2006, 69(4):260-266.
[3]	M¨AKITALO M, FOI A. Optimal inversion of theAnscombe transformation in low-count Poisson imagedenoising [J]. IEEE Transactions on Image Processing,2011, 20(1): 99-109.
[4]	BUADES A, COLL B, MOREL J M. A review of imagedenoising algorithms, with a new one [J]. SIAMJournal on Multiscale Modeling & Simulation, 2005,4(2): 490-530.
[5]	FENG J Z, SONG L, HUO X M, et al. An optimizedpixel-wise weighting approach for patch-based imagedenoising [J]. IEEE Signal Processing Letters, 2014,22(1): 115-119.
[6]	DABOV K, FOI A, KATKOVNIK V, et al. Image denoisingby sparse 3D transform-domain collaborativefiltering [J]. IEEE Transactions on Image Processing,2007, 16(8): 2080-2095.
[7]	DELEDALLE C A, LTCI C, SALMON J, et al. Imagedenoising with patch based PCA: Local versus global[C]//Proceedings of the 22nd British Machine VisionConference. Dundee: BMVA Press, 2011: 1-10.
[8]	SALMON J, HARMANY Z, DELEDALLE C A, et al.Poisson noise reduction with non-local PCA [J]. Journalof Mathematical Imaging & Vision, 2014, 48(2):279-294.
[9]	GIRYES R, ELAD M. Sparsity-based poisson denoisingwith dictionary learning [J]. IEEE Transactions onImage Processing, 2014, 23(12): 5057-5069.
[10]	ZHANG X, BURGER M, BRESSON X, et al. Bregmanizednonlocal regularization for deconvolution andsparse reconstruction [J]. SIAM Journal on ImagingSciences, 2010, 3(3): 253-276.
[11]	LANDI G, PICCOLOMINI E L. NPTool: A Matlabsoftware for nonnegative image restoration withNewton projection methods [J]. Numerical Algorithms,2013, 62(3): 487-504.
[12]	LANDI G, PICCOLOMINI E L. An efficient methodfor nonnegatively constrained total variation-based denoisingof medical images corrupted by poisson noise[J]. Computerized Medical Imaging & Graphics, 2012,36(1): 38-46.
[13]	ZHANG Z R, HUANG L L, FEI X, et al. Image poissondenoising model and algorithm based on nonlocal TVregularization [J]. Journal of System Simulation, 2014,26(9): 2010-2015 (in Chinese).
[14]	CHAN T F, SUNG H K, SHEN J. Euler’s elastica andcurvature based inpaintings [J]. SIAM Journal of AppliedMathematics, 2002, 63(2): 564-592.
[15]	TAI X C, HAHN J, CHUNG G J. A fast algorithmfor Euler’s elastica model using augmented Lagrangianmethod [J]. SIAM Journal on Imaging Sciences, 2011,4(1): 313-344.
[16]	DUAN Y P, WANG Y, HAHN J. A fast augmentedLagrangian method for Euler’s elastica models [J]. NumericalMathematics Theory Methods & Applications,2013, 6(1): 47-71.
[17]	ECKSTEIN J, BERTSEKAS D P. On the Douglas-Rachford splitting method and the proximal point algorithmfor maximal monotone operators [J]. MathematicalProgramming, 1992, 55: 293-318.

Poisson Noise Removal Based on Nonlocal Total Variation with Euler’s Elastica Pre-processing

Poisson Noise Removal Based on Nonlocal Total Variation with Euler’s Elastica Pre-processing

Knowledge

Abstract

Cite this article

share this article

References 17

Related Articles 1

Recommended Articles

Metrics

Comments