计算机科学 ›› 2020, Vol. 47 ›› Issue (7): 130-134.doi: 10.11896/jsjkx.190600120

• 计算机图形学&多媒体 • 上一篇    下一篇

三维块匹配波域调和滤波图像去噪

吴静1, 周先春1,2, 徐新菊1, 黄金1   

  1. 1 南京信息工程大学电子与信息工程学院 南京210044
    2 南京信息工程大学江苏省大气环境与装备技术协同创新中心 南京210044
  • 收稿日期:2019-06-24 出版日期:2020-07-15 发布日期:2020-07-16
  • 通讯作者: 周先春(zhouxc2008@163.com)
  • 作者简介:001398@nuist.edu.cn
  • 基金资助:
    国家自然科学基金项目(11202106,61302188);江苏省“信息与通信工程”优势学科建设项目;江苏高校品牌专业建设工程资助项目

Image Denoising by Mixing 3D Block Matching with Harmonic Filtering in Transform Domain

WU Jing1, ZHOU Xian-chun1,2, XU Xin-ju1, HUANG Jin1   

  1. 1 School of Electronic and Information Engineering,Nanjing University of Information Science and Technology,Nanjing 210044,China
    2 Jiangsu Collaborative Innovation Center of Atmospheric Environment and Equipment Technology,Nanjing University of Information Science and Technology,Nanjing 210044,China
  • Received:2019-06-24 Online:2020-07-15 Published:2020-07-16
  • About author:WU Jing,born in 1997,postgraduate.Her main research interest is image processing.
    ZHOU Xian-chun,born in 1974,Ph.D,postgraduate supervisor.His main research interests include signal & information processing and image proces-sing.
  • Supported by:
    This work was supported by the National Natural Science Foundation of China (11202106,61302188), Jiangsu Province “Information and Communication Engineering” Advantage Discipline Construction Project and Jiangsu University Brand Professional Construction Project Support Project

摘要: 针对当前图像去噪算法缺乏对整体结构的分析以及运算量过大的不足,提出了一种利用波域调和滤波扩散模型改进BM3D去噪技术的新算法。首先,利用传统的欧氏距离法将相似二维图像块合并,得到三维数组,再将联合滤波后的三维数组进行逆变换,得到图像的预估计数据。其次,通过小波分解变换提取预估计图像中的高频部分进行滤波,为避免边缘模糊,引用拉普拉斯高斯算法构建新算子并将其代入扩散模型。最后,进行小波重构,以得到原始图像的最终逼近,从而均衡运算速度和去噪性能,保护图像完整的结构信息。实验结果表明,新算法的去噪性能优异,内部信息保护更具完整性,运算速度合理,有利于实际应用。

关键词: 拉普拉斯高斯算法, 三维块匹配, 图像去噪, 小波分解

Abstract: Aiming at remedy the defeat that the current denoising algorithms lack of analyses of integral structure and excessive computational complexity,this paper proposes an improved denoising algorithm using the harmonic filtering diffusion model in wave-domain to amend BM3D technology.Firstly,the algorithm uses the Euclidean distance to merge similar 2-D image fragments thus obtaining 3-D date arrays.Then it is dealt by collaborative filtering,and the pre-estimation data of the image would be obtained by inverse 3-D transformation.Wavelet decomposition is used to extract high frequency part of pre-denoised image to filter.Lastly,wavelet reconstruction is conducted to estimate the original image,in order to avoid edge ambiguity.Laplacian of Gassian is used to construct a new operator into the diffusion model for filtering,so as to balance the operation speed and denoising performance,and protect the complete structureof the image information.The experimental results show that the new algorithm has perfect denoising performance,more integrity of internal information protection,and short running time,which is beneficial to practical applications.

Key words: BM3D, Image denoise, Laplacian of Gassian, Wavelet decomposition

中图分类号: 

  • TP391
[1]WU J B,YIN Z P,XIONG Y L.The fast multilevel fuzzy edge detection of blurry images[J].Ieee Signal Processing Letters,2007,14(5):344-347.
[2]LENG X G,JI K F,XING X W,et al.Hybrid bilateral filtering algorithm based on edge detection[J].IET Image Processing,2016,10(11):809-816.
[3]CHEN J W,JIAO L C,MA W P.Unsupervised High-Level Feature Extraction of SAR Imagery With Structured Sparsity Priors and Incremental Dictionary Learning[J].IEEE Geoscience and Remote Sensing Letters,2016,13(10):1467-1471.
[4]OMAR C R,PATRICIA S,KRIKOR B O.Temporal Pattern Recognition in Gait Activities Recorded With a Footprint Imaging Sensor System[J].IEEE Sensors Journal,2016,16(24):8815-8822.
[5]冈萨雷斯RC,伍兹.RE.数字图像处理(第三版)[M].阮秋琦,阮宇智,等译.北京:电子工业出版社,2010:197-213.
[6]WANG Z,HUANG X,LI Y X,et al.A new image encryption algorithm based on the fractional-order hyperchaotic Lorenz system[J].China Phys B,2013,22(1):010504.
[7]ZHOU X C,WANG M L,ZHOU L F.Image smoothing algorithm based on matching normal distribution diffusion[J].Journal of Image and Graph,2015,20(2):169-176.
[8]ZHOU X C,WANG M L,SHI L F,et al.Image smoothing model based on the combination of the gradient and curvature [J].Acta Physica Sinica,2015,64(4):044201(1-7).
[9]BUADES A,COLL B,MOREL J M.A non-local algorithm for image denoising[J].IEEE Computer Vision and Pattern Recognition,2005(1):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]LI Y J,ZHANG J W,WANG M M.Improved BM3D denoising method[J].IET Image Processing,2017,11(12):1197-1204.
[12]ISIDORA S,IGOR D,MILOS D.Adaptive average BM3D filter for resconstruction of images with combined noise[C]//2018 7th Mediterranean Conference on Embedded Computing(MECO).2018:1-4.
[13]ZHOU X C,WANG M L,SHI L F,et al.Diffusion denoisingmodel based on the wavelet and biharmonic equation[J].Acta Physica Sinica,2015,64(6):64203.
[14]FENG X C,LI X H,WANG W W.Improvement of BM3D Algorithm Based on Wavelet and Directed Diffusion[C]//IEEE 2017 International Conference on Machine Vision and Information Technology (CMVIT).2017.
[15]ZHOU D,CHENG W.Image denoising with an optimal threshold and neighbouring window[J].Pattern Recognition Letters,2008,29:1694-1697.
[16]ALESSANDRO F,VLADMIR K,KAREN E.Pointwise Shape-Adaptive DCT for High-Quality Denoising and Deblocking of Grayscale and Color Images[J].IEEE Transactions on Image Processing,2007,16(5):1395-1411.
[17]ABDERRAHIM E,XAVIER D,OLIVIER L.Non-Local Morphological PDEs and ρ-Laplacian Equation on Graphs With Applications in Image Processing and Machine Learning[J].IEEE Journal of Selected Topics in Signal Processing,2012,6(7):764-779.
[18]MALLAT S,HWANG W L.IEEE Trans on IT 38612.Mallat S Hwang W L.Singulatitr Detection and Processing with Wavelet[J].IEEE Transactions on IT,1992,38(2):612-643.
[19]PERONA P,MALIK J.Scale-spaceand edge detection using anisotropic diffusion[J].IEEE Transactions on Pattern Analysis & Machine Intelligence,1990,12(7):629-639.
[1] 钟岳, 方虎生, 张国玉, 王钊, 朱经纬.
基于9轴姿态传感器的CNN旗语动作识别方法
Method of CNN Flag Movement Recognition Based on 9-axis Attitude Sensor
计算机科学, 2021, 48(6): 153-158. https://doi.org/10.11896/jsjkx.200500005
[2] 巫勇, 刘永坚, 唐瑭, 王洪林, 郑建成.
基于鲁棒低秩张量恢复的高光谱图像去噪
Hyperspectral Image Denoising Based on Robust Low Rank Tensor Restoration
计算机科学, 2021, 48(11A): 303-307. https://doi.org/10.11896/jsjkx.210200103
[3] 曹义亲, 谢舒慧.
基于网格搜索的特定类别图像去噪算法
Category-specific Image Denoising Algorithm Based on Grid Search
计算机科学, 2020, 47(11): 168-173. https://doi.org/10.11896/jsjkx.190900004
[4] 李桂会,李晋江,范辉.
自适应匹配追踪图像去噪算法
Image Denoising Algorithm Based on Adaptive Matching Pursuit
计算机科学, 2020, 47(1): 176-185. https://doi.org/10.11896/jsjkx.181202280
[5] 肖佳, 张俊华, 梅礼晔.
改进的三维块匹配去噪算法
Improved Block-matching 3D Denoising Algorithm
计算机科学, 2019, 46(6): 288-294. https://doi.org/10.11896/j.issn.1002-137X.2019.06.043
[6] 刘佩, 贾建, 陈莉, 安影.
基于快速自适应的二维经验模态分解的图像去噪算法
Image Denoising Algorithm Based on Fast and Adaptive Bidimensional Empirical Mode Decomposition
计算机科学, 2019, 46(11): 260-266. https://doi.org/10.11896/jsjkx.190400159
[7] 张真真,王建林.
结合第二代Bandelet变换分块的字典学习图像去噪算法
Dictionary Learning Image Denoising Algorithm Combining Second Generation Bandelet Transform Block
计算机科学, 2018, 45(7): 264-270. https://doi.org/10.11896/j.issn.1002-137X.2018.07.046
[8] 赵利博,刘奇,付方玲,何凌.
基于小波变换和倒谱分析的腭裂高鼻音等级自动识别
Automatic Detection of Hypernasality Grades Based on Discrete Wavelet Transformation and Cepstrum Analysis
计算机科学, 2018, 45(4): 278-284. https://doi.org/10.11896/j.issn.1002-137X.2018.04.047
[9] 焦莉娟,王文剑.
一种基于差异系数的稀疏度自适应图像去噪算法
Sparsity-adaptive Image Denoising Algorithm Based on Difference Coefficient
计算机科学, 2018, 45(2): 94-97. https://doi.org/10.11896/j.issn.1002-137X.2018.02.016
[10] 赵杰,马玉娇,刘帅奇.
结合视觉显著性的图像去噪优化算法
Image Denoising Optimization Algorithm Combined with Visual Saliency
计算机科学, 2018, 45(2): 312-317. https://doi.org/10.11896/j.issn.1002-137X.2018.02.054
[11] 陈鹏, 张建伟.
结合核函数与非线性偏微分方程的图像去噪方法
Image Denoising Method Combining Kernel Function and Nonlinear Partial Differential Equation
计算机科学, 2018, 45(11): 278-282. https://doi.org/10.11896/j.issn.1002-137X.2018.11.044
[12] 马洪晋, 聂玉峰.
基于二级修复的多方向加权均值滤波算法
Multi-directional Weighted Mean Denoising Algorithm Based on Two Stage Noise Restoration
计算机科学, 2018, 45(10): 250-254. https://doi.org/10.11896/j.issn.1002-137X.2018.10.046
[13] 赵杰,王配配,门国尊.
基于非局部相似和低秩矩阵逼近的SAR图像去噪
SAR Image Denosing Based on Nonlocal Similarity and Low Rank Matrix Approximation
计算机科学, 2017, 44(Z6): 183-187. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.042
[14] 孙少超.
一种非凸核范数最小化一般模型及其在图像去噪中的应用
Nonconvex Muclear Morm Minimization General Model with Its Application in Image Denoising
计算机科学, 2017, 44(Z6): 236-239. https://doi.org/10.11896/j.issn.1002-137X.2017.6A.054
[15] 张爱玲,李鹏,刘晟.
基于粒子群算法的图像椒盐噪声去除算法
Algorithm of Image Salt and Pepper Noise Elimination Based on Particle Swarm Algorithm
计算机科学, 2017, 44(8): 301-305. https://doi.org/10.11896/j.issn.1002-137X.2017.08.052
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
No Suggested Reading articles found!