自适应匹配追踪图像去噪算法

doi:10.11896/jsjkx.181202280

摘要/Abstract

摘要： 针对目前的稀疏去噪算法分解效率低、去噪效果不理想的问题,提出了一种基于自适应匹配追踪的图像去噪算法。该算法首先通过自适应匹配追踪算法求解稀疏系数,然后利用K奇异值分解算法将字典训练成能够有效反映图像结构特征的自适应字典,最后将稀疏系数与自适应字典相结合来重构图像。在重构过程中,将噪声对应的系数去除,最终达到去噪的效果。算法引入Spike-Slab先验来引导稀疏系数矩阵的稀疏性,并利用两个权重矩阵促使去噪模型更加真实。鉴于字典在稀疏算法中的重要性,将自适应字典与DCT冗余字典、Global字典进行比较。实验结果显示,选择自适应字典的去噪结果比传统字典在峰值信噪比上高出约4.5dB;与目前6种主流的稀疏去噪方法相比,文中提出的方法在3种评价指标上均有不同程度的提高,其中峰值信噪比平均提高了约0.76~6.24dB,特征相似度平均提高了约0.012~0.082,结构相似性平均提高了约0.015~0.108。对图像去噪算法进行定性的评价,结果显示所提算法保留了更多的有用信息,视觉效果最佳。实验充分证明了自适应匹配追踪图像去噪算法对图像去噪的有效性和鲁棒性。

关键词: K奇异值分解, Spike-Slab先验, 图像去噪, 稀疏表示, 自适应匹配追踪

Abstract: Aiming at the problem that the current sparse denoising algorithm has low decomposition efficiency and unsatisfactory denoising effect,an image denoising algorithm based on adaptive matching pursuit was proposed.Firstly,the algorithm uses the adaptive matching pursuit algorithm to solve the sparse coefficients,and then uses the K-means singular value decomposition algorithm to train the dictionary into an adaptive dictionary that can effectively reflect the image structure features.Finally,theima-ge is reconstructed by combining the sparse coefficient with the adaptive dictionary.During the reconstruction process,the coefficients corresponding to the noise are removed,and finally the denoising effect is achieved.Spike-Slab priori is introduced to guide the sparsity of sparse coefficient matrix,and two weight matrices are used to make the denoising model more realistic.In view of the importance of dictionary in sparse algorithm,this paper compared adaptive dictionary with DCT redundant dictionary and Global dictionary.The experimental results show that the denoising result of adaptive dictionary is about 4.5 dB higher than that of traditional dictionary in terms of peak signal-to-noise ratio (PSNR).The proposed method improves three evaluation indicators in varying degrees compared with the current six main methods of sparse denoising.The PSNR is increased by about 0.76dB to 6.24 dB,the feature similarity (FSIM) is increased by about 0.012 to 0.082,and the structure similarity (SSIM) is increased by about 0.015 to 0.108 on average.The qualitative evaluation of the image denoising algorithm shows that the proposed algorithm retains more useful information and has the best visual effect.Therefore,the experiment fully proves its effectiveness and robustness.

Key words: Adaptive matching pursuit, Image denoising, K-means singular value decomposition, Sparse representation, Spike-Slab priori

中图分类号:

TP391.41

李桂会,李晋江,范辉. 自适应匹配追踪图像去噪算法[J]. 计算机科学, 2020, 47(1): 176-185. https://doi.org/10.11896/jsjkx.181202280

LI Gui-hui,LI Jin-jiang,FAN Hui. Image Denoising Algorithm Based on Adaptive Matching Pursuit[J]. Computer Science, 2020, 47(1): 176-185. https://doi.org/10.11896/jsjkx.181202280

参考文献

[1]JUBAIR I,RAHMAN M,ASHFAQUEUDDIN S,et al.An enhanced decision based adaptive median filtering technique to remove Salt and Pepper noise in digital images[C]∥International Conference on Computer & Information Technology.Dhaka:IEEE press,2011:428-433.
[2]NAVEED K,SHAUKAT B,REHMAN N U.Signal denoising based on dual tree complex wavelet transform and goodness of fit test[C]∥International Conference on Digital Signal Proces-sing.London:IEEE,2017:1-5.
[3]YAGAN A C,OZGEN M T.A spectral graph wiener filter in graph fourier domain for improved image denoising[C]∥2016 IEEE Global Conference on Signal and Information Processing (GlobalSIP).Washington:IEEE,2016:450-454.
[4]DAS S L,NACHIAPPAN A.Role of hybrid switching filter in image denoising - a comparative study[C]∥2012 Annual IEEE India Conference (INDICON).Kochi:IEEE,2012:1180-1183.
[5]CAO Y,LUO Y P,YANG S Y.Hybrid Linear Model Based Ima- ge Denoising[J].Chinese Journal of Computers,2009,32(11):2260-2264.
[6]RAJA H,BAJWA W U.Cloud K-SVD:A Collaborative Dictionary Learning Algorithm for Big,Distributed Data[J].IEEE Transactions on Signal Processing,2016,64(1):173-188.
[7]JIA L N,SONG S T,YAO L H,et al.Image Denoising via Sparse Representation over Grouped Dictionaries with Adaptive Atom Size[J].IEEE Access,2017,5:22514-22529.
[8]LU C,SHI J,JIA J.Scale Adaptive Dictionary Learning[J]. IEEE Transactions on Image Processing,2014,23(2):837-847.
[9]ROMANO Y,ELAD M.Patch-Disagreement as a Way to Improve K-SVD Denoising[C]∥IEEE International Conference on Acoustics.Brisbane:IEEE,2015:1280-1284.
[10]ENGAN K,AASE S O ,HUSOY J H.Method of optimal directions for frame design[C]∥IEEE International Conference on Acoustics,Speech,and Signal Processing.USA:IEEE,1999:2443-2446.
[11]AHARON M,ELAD M,BRUCKSTEIN A.MYMrm KMYM-SVD:An Algorithm for Designing Overcomplete Dictionaries for Sparse Representation[J].IEEE Transactions on Signal Processing,2006,54(11):4311-4322.
[12]MALLAT S G,ZHANG Z.Matching pursuits with time-frequency dictionaries[J].IEEE Transactions on Signal Processing,1993,41(12):3397-3415.
[13]TROPP J A,GILBERT A C.Signal Recovery From Random Measurements Via Orthogonal Matching Pursuit[J].IEEE Transactions on Information Theory,2007,53(12):4655-4666.
[14]DONOHO D L,TSAIG Y,DRORI I,et al.Sparse Solution of Underdetermined Systems of Linear Equations by Stagewise Orthogonal Matching Pursuit[J].IEEE Transactions on Information Theory,2012,58(2):1094-1121.
[15]TANG L,CHEN M J.Image Denoising Method Using the Gradient Matching Pursuit[J].Mathematical Modelling of Engineering Problems,2016,3(2):53-56.
[16]DENG X,LIU Z.Image denoising based on steepest descent OMP and K-SVD[C]∥IEEE International Conference on Signal Processing.Ningbo:IEEE,2015:1-5.
[17]LI S,FANG L.Signal Denoising With Random Refined Orthog- onal Matching Pursuit[J].IEEE Transactions on Instrumentation and Measurement,2012,61(1):26-34.
[18]DO T T,GAN L,NGUYEN N,et al.Sparsity adaptive matching pursuit algorithm for practical compressed sensing[C]∥Conference on Signals,Systems & Computers.Pacific Grove:IEEE,2008:581-587.
[19]YUAN S,WANG S,MA M,et al.Sparse Bayesian Learning-Based Time-Variant Deconvolution[J].IEEE Transactions on Geoscience & Remote Sensing,2017,55(11):6182-6194.
[20]JIN M,ROTH S,FAVARO P.Noise-Blind Image Deblurring
[C]∥IEEE Conference on Computer Vision & Pattern Recognition.Honolulu:IEEE,2017:3834-3842.
[21]VU T H,MOUSAVI H S,MONGA V.Adaptive matching pursuit for sparse signal recovery[C]∥IEEE International Conference on Acoustics.New Orleans:IEEE,2017:4331-4335.
[22]OLIVA G,SETOLA R,HADJICOSTIS C N.Distributed asynchronous Cholesky decomposition[C]∥Decision & Control.Las Vegas:IEEE,2016:4414-4419.
[23]MITCHELL T J,BEAUCHAMP J J.Bayesian Variable Selection in Linear Regression [J].Publications of the American Statistical Association,1988,83(404):1023-1032.
[24]GEORGE E,MCCULLOCH R.Variable Selection via Gibbs Sampling[J].Publications of the American Statistical Association,1993,88(423):881-889.
[25]CHEN B,PAISLEY J W,CARIN L.Sparse linear regression with beta process priors[C]∥IEEE International Conference on Acoustics Speech & Signal Processing.Dallas:IEEE,2010:1234-1237.
[26]ZHUANG P X,HUANG Y,ZENG D L,et al.Mixed noise removal based on a novel non-parametric Bayesian sparse outlier model[J].Neurocomputing,2016,174(PB):858-865.
[27]DING X H,MI Z Y,HUANG Y,et al.Robust rvm based on spike-slab prior[J].Journal of Electronics (China),2012,29(6):593-597.
[28]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.
[29]ZHANG L,ZHANG L,MOU X Q,et al.FSIM:A Feature Similarity Index for Image Quality Assessment[J].IEEE Transactions on Image Processing,2011,20(8):2378-2386.

相关文章 15

[1]	李培冠, 於志勇, 黄昉菀. 基于稀疏表示的电力负荷数据补全 Power Load Data Completion Based on Sparse Representation 计算机科学, 2021, 48(2): 128-133. https://doi.org/10.11896/jsjkx.191200152
[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]	田旭, 常侃, 黄升, 覃团发. 基于残差字典及协作表达的单图像超分辨率算法 Single Image Super-resolution Algorithm Using Residual Dictionary and Collaborative Representation 计算机科学, 2020, 47(9): 135-141. https://doi.org/10.11896/jsjkx.190600146
[4]	吴静, 周先春, 徐新菊, 黄金. 三维块匹配波域调和滤波图像去噪 Image Denoising by Mixing 3D Block Matching with Harmonic Filtering in Transform Domain 计算机科学, 2020, 47(7): 130-134. https://doi.org/10.11896/jsjkx.190600120
[5]	程中建, 周双娥, 李康. 基于多尺度自适应权重的稀疏表示目标跟踪算法 Sparse Representation Target Tracking Algorithm Based on Multi-scale Adaptive Weight 计算机科学, 2020, 47(6A): 181-186. https://doi.org/10.11896/JsJkx.190500093
[6]	吴庆洪, 高晓东. 稀疏表示和支持向量机相融合的非理想环境人脸识别 Face Recognition in Non-ideal Environment Based on Sparse Representation and Support Vector Machine 计算机科学, 2020, 47(6): 121-125. https://doi.org/10.11896/jsjkx.190500058
[7]	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
[8]	李笑宇,高清维,卢一相,孙冬. 一种根据图像能量调整的图像融合方法 Image Fusion Method Based on Image Energy Adjustment 计算机科学, 2020, 47(1): 153-158. https://doi.org/10.11896/jsjkx.181202437
[9]	张冰, 谢从华, 刘哲. 基于显著稀疏表示和邻域信息的多聚焦图像融合 Multi-focus Image Fusion Based on Latent Sparse Representation and Neighborhood Information 计算机科学, 2019, 46(9): 254-258. https://doi.org/10.11896/j.issn.1002-137X.2019.09.038
[10]	宋晓祥,郭艳,李宁,余东平. 基于稀疏贝叶斯学习的协同进化时间序列缺失数据预测算法 Missing Data Prediction Algorithm Based on Sparse Bayesian Learning in Coevolving Time Series 计算机科学, 2019, 46(7): 217-223. https://doi.org/10.11896/j.issn.1002-137X.2019.07.033
[11]	王丽芳, 史超宇, 蔺素珍, 秦品乐, 高媛. 基于联合图像块聚类自适应字典学习的多模态医学图像融合 Multi-modal Medical Image Fusion Based on Joint Patch Clustering of Adaptive Dictionary Learning 计算机科学, 2019, 46(7): 238-245. https://doi.org/10.11896/j.issn.1002-137X.2019.07.036
[12]	张福旺, 苑会娟. 一种自适应稀疏表示和非局部自相似性的图像超分辨率重建算法 Image Super-resolution Reconstruction Algorithm with Adaptive Sparse Representationand Non-local Self-similarity 计算机科学, 2019, 46(6A): 188-191.
[13]	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
[14]	杜秀丽, 左思铭, 邱少明. 基于图像灰度熵的自适应字典学习算法 Adaptive Dictionary Learning Algorithm Based on Image Gray Entropy 计算机科学, 2019, 46(5): 266-271. https://doi.org/10.11896/j.issn.1002-137X.2019.05.041
[15]	茹锋, 徐锦, 常琪, 阚丹会. 一种用于影像遗传学关联分析的高阶统计量结构化稀疏算法 High Order Statistics Structured Sparse Algorithm for Image Genetic Association Analysis 计算机科学, 2019, 46(4): 66-72. https://doi.org/10.11896/j.issn.1002-137X.2019.04.010

Metrics

Viewed

Full text

Abstract

Cited

Shared

Discussed