自适应非局部数据保真项和双边总变分的图像去噪模型

doi:10.11772/j.issn.1001-9081.2017.08.2334

计算机应用 ›› 2017, Vol. 37 ›› Issue (8): 2334-2342.DOI: 10.11772/j.issn.1001-9081.2017.08.2334

• 计算机视觉与虚拟现实 • 上一篇下一篇

自适应非局部数据保真项和双边总变分的图像去噪模型

郭黎^1,2, 廖宇¹, 李敏¹, 袁海林¹, 李军¹

1. 湖北民族学院信息工程学院, 湖北恩施 445000;
2. University of Groningen, Faculty of Mathematics and Computer Science, Groningen the Netherlands, 9747AG

收稿日期:2017-01-13 修回日期:2017-03-20 出版日期:2017-08-10 发布日期:2017-08-12
通讯作者: 郭黎
作者简介:郭黎(1978-),女,湖北黄冈人,副教授,博士,主要研究方向:图像处理与识别、计算机视觉、信息显示;廖宇(1979-),男,湖北建始人,副教授,博士研究生,主要研究方向:信号与信息处理、光通信;李敏(1981-),女,湖北宣恩人,讲师,硕士,主要研究方向:信号与信息处理;袁海林(1967-),男,湖北恩施人,教授,硕士,主要研究方向:嵌入式系统分析;李军(1971-),男,湖北利川人,教授,硕士,主要研究方向:计算机图形学、数字图像处理。
基金资助:
国家自然科学基金资助项目（61663008，61463014，61562025）；国家科技支撑计划项目（2015BAK27B01）；湖北省自然科学基金资助项目（2015CFC781，2014CFB612，2012FFC02601）；四川省教育厅科研项目（15ZB0039）；国家留学基金委地方合作项目；湖北民族学院博士启动基金资助项目（MY2014B018）。

Image denoising model with adaptive non-local data-fidelity term and bilateral total variation

GUO Li^1,2, LIAO Yu¹, LI Min¹, YUAN Hailin¹, LI Jun¹

1. College of Information Engineering, Hubei University for Nationalities, Enshi Hubei 445000, China;
2. Faculty of Mathematics and Computer Science, University of Groningen, Groningen 9747 AG, the Netherlands

Received:2017-01-13 Revised:2017-03-20 Online:2017-08-10 Published:2017-08-12
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61663008,61463014,61562025),the National Key Technology R&D Program (2015BAK27B01),the Natural Science Foundation of Hubei Province (2015CFC781,2014CFB612,2012FFC02601),the Scientific Research Project of Department of Education of Sichuan Province (15ZB0039),the Foundation of China Scholarship Council,the Foundation of PhD Technology Project of Hubei University for Nationalities (MY2014B018).

摘要/Abstract

摘要： 针对常见去噪方法容易造成特定区域过度平滑、奇异结构残余噪声以及产生阶梯效应和对比度损失等问题，提出一种自适应非局部数据保真项和双边总变分的图像去噪模型，建立了自适应非局部正则化能量泛函和相应的变分框架。首先，对噪声图像利用自适应权值的非局部均值求得数据拟合项；其次，引入双边总变分正则化项，利用正则化系数来适度平衡数据拟合项和正则化项的影响；最后，通过能量函数最小化对不同的噪声统计快速求得最优解，从而达到降低残余噪声并纠正过度平滑的目的。通过理论分析和针对模拟噪声图像与真实噪声图像的实验结果表明，所提出的图像去噪模型能够较好地处理具有不同统计特性的图像噪声，与自适应非局部均值滤波去噪相比，所提算法的峰值信噪比（PSNR）值最多可以得到0.6 dB的改善；与全变分正则化图像去噪算法比较，所提算法的主观视觉效果明显更好，在去噪的同时图像纹理和边缘等细节信息保护得更好，PSNR值最多可以提高10 dB，而多尺度结构相似性度（MS-SSIM）指标可以提升0.3。因此，所提出的图像去噪模型可以在理论上更好地探讨如何合理处理噪声和图像内容本身的高频细节信息，在视频和图像分辨率提升等领域也具有良好的实际应用价值。

关键词: 自适应非局部均值, 数据保真项, 正则化函数, 双边总变分, 图像去噪

Abstract: Aiming at the problems of over-smoothing, singular structure residual noise, contrast loss and stair effect of common denoising methods, an image denoising model with adaptive non-local data fidelity and bilateral total variation regularization was proposed, which provides an adaptive non-local regularization energy function and the corresponding variation framework. Firstly, the data fidelity term was obtained by non-local means filter with adaptive weighting method. Secondly, the bilateral total variation regularization was introduced in this framework, and a regularization factor was used to balance the data fidelity term and the regularization term. At last, the optimal solutions for different noise statistics were obtained by minimizing the energy function, so as to achieve the purpose of reducing residual noise and correcting excessive smoothing. The theoretical analysis and simulation results on simulated noise images and real noise images show that the proposed image denoising model can deal with different statistical noise in image, and the Peak-Signal-to-Noise Ratio (PSNR) of it can be increased by up to 0.6 dB when compared with the adaptive non-local means filter; when compared with the total variation regularization algorithm, the subjective visual effect of the proposed model was improved obviously and the details of image texture and edges was protected very well when denoising, and the PSNR was increased by up to 10 dB, the Multi-Scale Structural Similarity index (MS-SSIM) was increased by 0.3. Therefore, the proposed denoising model can theoretically better deal with the noise and the high frequency detail information of the image, and has good practical application value in the fields of video and image resolution enhancement.

Key words: adaptive non-local means, data fidelity term, regularization function, bilateral total variation, image denoising

中图分类号:

郭黎, 廖宇, 李敏, 袁海林, 李军. 自适应非局部数据保真项和双边总变分的图像去噪模型[J]. 计算机应用, 2017, 37(8): 2334-2342.

GUO Li, LIAO Yu, LI Min, YUAN Hailin, LI Jun. Image denoising model with adaptive non-local data-fidelity term and bilateral total variation[J]. Journal of Computer Applications, 2017, 37(8): 2334-2342.

参考文献

[1] 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 Z R, HUANG L L, FEI X, et al. Image Possion denoising model and algorithm based on nonlocal TV regularization[J]. Journal of System Simulation, 2014, 26(9):2110-2115.)
[2] 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 X G, LI Y. Improved image Possion denoising model based on fractional variation[J]. Journal of Computer Applications, 2013, 33(4):1100-1102.)
[3] BUADES A, COLL B, MOREL J-M. A review of image denoising algorithm, with a new one[J]. SIAM Journal on Multiscale Modeling and Simulation:A SIAM Interdisciplinary Journal, 2005, 4(2):490-530.
[4] BUADES A, COLL B, MOREL J-M. Non-local image and movie denoising[J]. International Journal of Computer Vision, 2007, 76(2):123-139.
[5] 杨爱萍,田玉珍,何宇清,等.基于改进K-SVD和非局部正则化的图像去噪[J].计算机工程,2015,41(5):249-253. (YANG A P, TIAN Y Z, HE Y Q, et al. Image denoising based on improved K-SVD and non-local regularization[J]. Computer Engineering, 2015, 5:249-253.)
[6] 费选,违志辉,肖亮,等.优化加权TV的复合正则化压缩感知图像重建[J].中国图象图形学报,2014,19(2):211-218. (FEI X, WEI Z H, XIAO L, et al. Compound regularized compressed sensing image reconstruction based on optimal reweighted TV[J]. Journal of Image and Graphics, 2014, 19(2):211-218.)
[7] 耿海, 何小卫, 樊骏笠. 基于可变指数及L1保真项的图像去噪算法[J].计算机应用,2013,33(10):2931-2934. (GENG H, HE X W, FAN J L. Image denoising algorithm with variable exponent regularization and L1 fidelity[J]. Journal of Computer Applications, 2013, 33(10):2931-2934.)
[8] 史加荣, 郑秀云,杨威.不完全鲁棒主成分分析的正则化方法及其在背景建模中的应用[J].计算机应用,2015,35(10):2824-2827. (SHI J R, ZHENG X Y, YANG W. Regularized approach for incomplete robust principal component analysis and its application in background modeling[J]. Journal of Computer Applications, 2015, 35(10):2824-2827.)
[9] 刘红丽,刘伟锋,王延江.Hessian正则化Logistic回归模型[J].计算机工程与应用,2016,52(5):236-240. (LIU H L, LIU W F, WANG Y J. Hessian regularized logistic regression[J]. Computer Engineering and Applications, 2016, 52(5):236-240.)
[10] 冀常鹏, 姬红红, 郭伟平. 变阶变正则化APA在回声抵消中的应用[J].计算机工程与应用,2014,50(15):168-171. (JI C P, JI H H, GUO W P. APA with evolving order and variable regularization for echo cancellation[J]. Computer Engineering and Applications, 2014, 50(15):168-171.)
[11] TIKHONOV A N, ARSENIN V Y. Solutions of ill-posed problems[J]. Mathematics and Computation, 1978, 32(144):1320-1322.
[12] RUDIN L I, OSHER S, FATEMI E. Nonlinear total variation based noise removal algorithms[J]. Physica D:Nolinear Phenomena, 1992, 60(1/2/3/4):259-268.
[13] SELESNICK I W, PAREKH A, BAYRAM I. Convex 1-D total variation denoising with non-convex regularization[J]. IEEE Signal Processing Letters, 2015, 22(2):141-144.
[14] SUTOUR C, DELEDALLE C-A, AUJOL J-F. Adaptive regularization of the NL-means:application to image and video denoising[J]. IEEE Transaction on Image Processing, 2014, 23(8):3506-3521.
[15] FARSIU S, ROBINSON D, ELAD M, et al. Fast and robust multiframe super resolution[J]. IEEE Transactions on Image Processing, 2004, 13(10):1327-1344.
[16] FARSIU S, ROBINSON D, ELAD M, et al. Robust shift and add approach to superresolution[C]//Proceeding of SPIE International Conference on Applications of Digital Signal and Image Processing XXVI, SPIE 5203. Bellingham, WA:SPIE, 2003:121-130.
[17] DELEDALLE C-A, DUVAL V, SALMON J. Non-Local Methods with Shape-Adaptive Patches (NLM-SAP)[J]. Journal of Mathematical Imaging and Vision, 2011, 43(2):103-120.
[18] DELEDALLE C A, DENIS L, TUPIN F, et al. NL-SAR:a unified non-local framework for resolution-preserving in SAR denoising[J]. IEEE Transaction on Geoscience and Remote Sensing, 2015, 53(4):2021-2038.
[19] GUO L, HE X H, CHEN W L, et al. A pioneering video-to-video super-resolution reconstruction algorithm based on segmentation and space-time regularization[J]. The Imaging Science Journal, 2014, 62(4):236-250.

自适应非局部数据保真项和双边总变分的图像去噪模型

Image denoising model with adaptive non-local data-fidelity term and bilateral total variation

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	陶永鹏, 景雨, 顼聪. 基于分组字典与变分模型的图像去噪算法[J]. 计算机应用, 2019, 39(2): 551-555.
[2]	张雯雯, 韩裕生. 基于非局部自相似性的低秩稀疏图像去噪[J]. 计算机应用, 2018, 38(9): 2696-2700.
[3]	徐苏, 周颖玥. 基于图像分割的非局部均值去噪算法[J]. 计算机应用, 2017, 37(7): 2078-2083.
[4]	彭政, 陈东方, 王晓峰. 正则化超分辨率重建过程的自适应阈值去噪[J]. 计算机应用, 2017, 37(7): 2084-2088.
[5]	孙少超. 基于带权核范数最小化和混合高斯模型的去噪模型[J]. 计算机应用, 2017, 37(5): 1471-1474.
[6]	张新明, 康强, 程金凤, 涂强. 采用自适应四点窗中点滤波的高椒盐噪声滤除方法[J]. 计算机应用, 2017, 37(3): 832-838.
[7]	周尚波, 王李平, 尹学辉. 分数阶偏微分方程在图像处理中的应用[J]. 计算机应用, 2017, 37(2): 546-552.
[8]	张新明, 程金凤, 康强, 王霞. 迭代自适应权重均值滤波的图像去噪[J]. 计算机应用, 2017, 37(11): 3168-3175.
[9]	余应淮, 谢仕义. 滤除椒盐噪声的开关核回归拟合算法[J]. 计算机应用, 2017, 37(10): 2921-2925.
[10]	赵明超, 陈智斌, 文有为. 基于GPU图像去噪总变分对偶模型的并行计算[J]. 计算机应用, 2016, 36(5): 1228-1231.
[11]	赵井坤, 周颖玥, 林茂松. 基于稀疏表示与非局部相似的图像去噪算法[J]. 计算机应用, 2016, 36(2): 551-555.
[12]	黄智, 付兴武, 刘万军. 混合相似性权重的非局部均值去噪算法[J]. 计算机应用, 2016, 36(2): 556-562.
[13]	许志良, 邓承志. 基于非局部自相似的Shearlet自适应收缩图像去噪[J]. 计算机应用, 2015, 35(1): 235-238.
[14]	蒋伟李小龙杨永琴张恒. 基于有理数阶微分的图像去噪新方法[J]. 计算机应用, 2014, 34(3): 801-805.
[15]	黄果陈庆利许黎门涛蒲亦非. 带边缘补偿的分数阶积分图像去噪算法[J]. 计算机应用, 2014, 34(10): 2957-2962.