Please wait a minute...
Journal of ZheJiang University (Engineering Science)  2020, Vol. 54 Issue (6): 1164-1169    DOI: 10.3785/j.issn.1008-973X.2020.06.013
Computer Technology     
Image Poisson denoising algorithm based on Markov fields of experts
Zhen JIA(),Wen-de DONG*(),Gui-li XU,Shi-peng ZHU
College of Automation Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 211106, China
Download: HTML     PDF(983KB) HTML
Export: BibTeX | EndNote (RIS)      

Abstract  

A Poisson noise image denoising method based on Bayesian probability model was proposed. An image denoising model was constructed based on Bayesian maximum a posteriori probability model and with combination of Poisson probability distribution. Considering that Markov random fields cannot represent complex natural images effectively, a higher-order Markov fields of experts was introduced as a prior regular term of the model to represent the probability distribution of the image. The quadratic penalty function was used to optimize the denoising model and restore clear images. The proposed method was compared with other denoising algorithms; the denoising effect was evaluated objectively by using two evaluation indexes: peak signal-to-noise ratio and structural similarity. The experimental results show that, compared with the traditional denoising methods, the peak signal-to-noise ratio of this method increased by at least 0.18 dB, and the denoising performance is significantly better than that of other methods. Thus, the details of the image can be retained better by using this mothed.



Key wordsPoisson noise      Markov fields of experts      regularization      image denoising      quadratic penalty function     
Received: 15 June 2019      Published: 06 July 2020
CLC:  TN 911.73  
Corresponding Authors: Wen-de DONG     E-mail: jiazhen_95@163.com;dongwende@nuaa.edu.cn
Cite this article:

Zhen JIA,Wen-de DONG,Gui-li XU,Shi-peng ZHU. Image Poisson denoising algorithm based on Markov fields of experts. Journal of ZheJiang University (Engineering Science), 2020, 54(6): 1164-1169.

URL:

http://www.zjujournals.com/eng/10.3785/j.issn.1008-973X.2020.06.013     OR     http://www.zjujournals.com/eng/Y2020/V54/I6/1164


基于马尔科夫专家场的泊松噪声图像去噪方法

提出一种基于贝叶斯概率模型的泊松噪声图像去噪方法. 该方法基于贝叶斯最大后验概率模型框架,结合泊松概率分布,构建图像去噪模型. 考虑到马尔科夫随机场不能对复杂自然图像有效表征,引入高阶的马尔科夫专家场作为模型先验正则项,以表征图像自身概率分布. 利用二次惩罚函数,优化求解去噪模型,还原清晰图像. 将所提方法与其他去噪算法进行仿真实验对比,并采用峰值信噪比和结构相似性2种评价指标对去噪效果进行客观评价. 实验结果表明:与传统去噪方法相比,该方法的峰值信噪比至少提升了0.18 dB,去噪性能显著优于其他方法,能更好地保留图像的细节信息.


关键词: 泊松噪声,  马尔科夫专家场,  正则化,  图像去噪,  二次惩罚函数 
Fig.1 Typical domain system for fields of experts(FoE)model
Fig.2 Quadratic penalty function error-convergence curve
Fig.3 Local nebula images of denoising results by nine denoising methods
Fig.4 Comparison of peak signal-noise-ratio of eight images with different denoising methods
Fig.5 Comparison of structural similarity of eight images with different denoising methods
[1]   张芳. 图像泊松去噪算法研究[D]. 杭州: 杭州电子科技大学, 2017: 10-11.
ZHANG Fang. Research on image Poisson denoising[D]. Hangzhou: Hangzhou Dianzi University, 2017: 10-11.
[2]   JIN X, XU Z, HIRAKAWA K Noise parameter estimation for Poisson corrupted images using variance stabilization transforms[J]. IEEE Transactions on Image Processing, 2014, 23 (3): 1329- 1339
doi: 10.1109/TIP.2014.2300813
[3]   王贵, 管志成 具有全局收敛性的彩色图像去噪模型[J]. 浙江大学学报: 工学版, 2005, (3): 78- 82
WANG Gui, GUAN Zhi-cheng Color image denoising model with global convergence[J]. Journal of Zhejiang University: Engineering Science, 2005, (3): 78- 82
[4]   BUADES A, COLL B, MOREL J M A review of image denoising algorithms, with a new one[J]. SIAM Journal on Multiscale Modeling and Simulation, 2005, 4 (2): 490- 530
doi: 10.1137/040616024
[5]   刘涛, 赵巨峰, 徐之海, 等 基于卡尔曼滤波的红外图像增强算法[J]. 浙江大学学报: 工学版, 2012, 46 (8): 1534- 1539
LIU Tao, ZHAO Ju-feng, XU Zhi-hai, et al Enhancement algorithm for infrared images based on Kalman filter[J]. Journal of Zhejiang University: Engineering Science, 2012, 46 (8): 1534- 1539
[6]   DEY N, BLANC F L, ZIMME R C, et al Richardson-Lucy algorithm with total variation regularization for 3D confocal microscope deconvolution[J]. Microscopy Research and Technique, 2010, 69 (4): 260- 266
[7]   张峥嵘, 刘红毅, 韦志辉 欧拉弹性正则化的图像泊松去噪[J]. 电子学报, 2017, 45 (1): 181- 191
ZHANG Zheng-rong, LIU Hong-yi, WEI Zhi-hui Image Poisson denoising based on Euler's elastica regularization[J]. Acta Electronica Sinica, 2017, 45 (1): 181- 191
doi: 10.3969/j.issn.0372-2112.2017.01.025
[8]   LE T, CHARTRAND R, ASAKI T J A variational approach to reconstructing images corrupted by Poisson noise[J]. Journal of Mathematical Imaging and Vision, 2007, 27 (3): 257- 263
doi: 10.1007/s10851-007-0652-y
[9]   GILBOA G, SOCHEN N, ZEEVI Y Y Variational denoising of partly textured images by spatially varying constraints[J]. IEEE Transactions on Image Processing A Publication of the IEEE Signal Processing Society, 2006, 15 (8): 2281
doi: 10.1109/TIP.2006.875247
[10]   张哲, 张化朋 一种基于偏微分方程变分去噪模型[J]. 计算机技术与发展, 2014, 24 (11): 103- 106
ZHANG Zhe, ZHANG Hua-peng A denoising model of variation based on PDE[J]. Computer Technology and Development, 2014, 24 (11): 103- 106
[11]   白键, 冯象初 一种基于积分微分方程的泊松噪声去除算法[J]. 电子与信息学报, 2013, 35 (2): 451- 456
BAI Jian, FENG Xiang-chu An integro-differential equation approach to reconstructing images corrupted by Poisson noise[J]. Journal of Electronics and Information Technology, 2013, 35 (2): 451- 456
[12]   胡学刚, 李妤 基于分数阶变分的图像泊松去噪模型[J]. 计算机应用, 2013, 33 (4): 1100- 1102
HU Xue-gang, LI Yu Improved image Poisson denoising model based on fractional variation[J]. Journal of Computer Applications, 2013, 33 (4): 1100- 1102
doi: 10.3724/SP.J.1087.2013.01100
[13]   张峥嵘, 黄丽丽, 费选, 等 非局部TV正则化的图像泊松去噪模型与算法[J]. 系统仿真学报, 2014, 26 (9): 2110- 2115
ZHANG Zheng-rong, HUANG Li-li, FEI Xuan, et al Image Poisson denoising model and algorithm based on nonlocal TV regularization[J]. Journal of System Simulation, 2014, 26 (9): 2110- 2115
[14]   王国权, 刘亮 FoE模型的训练方法研究[J]. 计算机技术与发展, 2010, 20 (12): 86- 89
WANG Guo-quan, LIU Liang A study on training methods of FoE model[J]. Computer Technology and Development, 2010, 20 (12): 86- 89
doi: 10.3969/j.issn.1673-629X.2010.12.022
[15]   张志, 王润生 边缘保持专家场模型的自然图像复原[J]. 自然科学进展, 2009, 19 (9): 1004- 1013
ZHANG Zhi, WANG Run-sheng Natural image restoration of the edge-preserving expert field model[J]. Natural Science Progress, 2009, 19 (9): 1004- 1013
doi: 10.3321/j.issn:1002-008X.2009.09.015
[16]   ROTH S, BLACK M J Fields of experts[J]. International Journal of Computer Vision, 2009, 82 (2): 205- 229
doi: 10.1007/s11263-008-0197-6
[17]   BESAG J E Spatial interaction and the statistical analysis of lattice systems[J]. Journal of the Royal Statistical Society. Series B: Methodological, 1974, 36 (2): 192- 236
[18]   KRISHNAN D, FERGUS R. Fast image deconvolution using Hyper-Laplacian priors [C] // Advances in Neural Information Processing Systems. Vancouver: NIPS, 2009: 1033-1041.
[19]   FIGUEIREDO M A T, BIOUCAS D J M. Deconvolution of Poissonian images using variable splitting and augmented Lagrangian optimization [C] // 2009 IEEE/SP 15th Workshop on Statistical Signal Processing. Cardiff: IEEE, 2009: 733-736.
[20]   TAO S, DONG W, FENG H, et al Non-blind image deconvolution using natural image gradient prior[J]. Optic-International Journal for Light and Electron Optics, 2013, 124 (24): 6599- 6605
doi: 10.1016/j.ijleo.2013.05.068
[21]   GU S, XIE Q, MENG D, et al Weighted nuclear norm minimization and its applications to low level vision[J]. International Journal of Computer Vision, 2017, 121 (2): 183- 208
doi: 10.1007/s11263-016-0930-5
[1] Xun CHENG,Jian-bo YU. Monitoring method for machining tool wear based on machine vision[J]. Journal of ZheJiang University (Engineering Science), 2021, 55(5): 896-904.
[2] YE Peng-zhao, FENG Hua-jun, XU Zhi-hai, LI Qi, CHEN Yue-ting. Blind restoration of compressed degraded image based onblock effect suppression[J]. Journal of ZheJiang University (Engineering Science), 2018, 52(2): 406-412.
[3] PAN Qiu-ping, WANG Zhen, GAN De-qiang, XIE Huan, LI Shang-yuan. Comparison of numerical differentiation based damping sensitivity method[J]. Journal of ZheJiang University (Engineering Science), 2018, 52(1): 174-183.
[4] CHEN Guo-xin, CHEN Sheng-chang. Regularization method with  p-norm sparsity constraints for potential field data reconstruction[J]. Journal of ZheJiang University (Engineering Science), 2014, 48(4): 748-785.
[5] PAN Jun, KONG Fan-sheng, WANG Rui-qin. Locality sensitive discriminant transductive learning[J]. Journal of ZheJiang University (Engineering Science), 2012, 46(6): 987-994.
[6] XIAO Dong-feng, YANG Chun-jie,SONG Zhi-huan. The forecasting model of blast furnace gas output
based on improved BP network
[J]. Journal of ZheJiang University (Engineering Science), 2012, 46(11): 2103-2108.
[7] QIU Guang, ZHENG Miao, ZHANG Hui, ZHU Jianke, BU Jia-jun, CHEN Chun, HANG Hang. Implicit product feature extraction through regularized topic modeling[J]. Journal of ZheJiang University (Engineering Science), 2011, 45(2): 288-294.