计算机科学 ›› 2019, Vol. 46 ›› Issue (5): 266-271.doi: 10.11896/j.issn.1002-137X.2019.05.041

• 图形图像与模式识别 • 上一篇    下一篇

基于图像灰度熵的自适应字典学习算法

杜秀丽, 左思铭, 邱少明   

  1. (大连大学通信与网络重点实验室 辽宁 大连116622)
    (大连大学信息工程学院 辽宁 大连116622)
  • 发布日期:2019-05-15
  • 作者简介:杜秀丽(1977-),女,博士,教授,CCF会员,主要研究方向为数字信号处理、通信技术,E-mail:22811623@qq.com(通信作者);左思铭(1993-),男,硕士,主要研究方向为数字信号处理;邱少明(1980-),男,硕士,副教授,主要研究方向为计算机技术与应用。
  • 基金资助:
    高速眼图测试关键技术研究基金(L2014495),辽宁“百千万人才工程”培养经费资助。

Adaptive Dictionary Learning Algorithm Based on Image Gray Entropy

DU Xiu-li, ZUO Si-ming, QIU Shao-ming   

  1. (Key Laboratory of Communication and Network,Dalian University,Dalian,Liaoning 116622,China)
    (College of Information Engineering,Dalian University,Dalian,Liaoning 116622,China)
  • Published:2019-05-15

摘要: 针对传统图像稀疏表示字典学习算法仅对图像训练学习单一字典,不能很好地对包含不同图像信息的图像块进行最优稀疏表示的问题,将图像灰度熵的思想引入到字典学习算法中,提出基于图像灰度熵的自适应字典学习算法。该算法将图像库作为训练样本,对图像库图像进行分块,计算各子块的灰度熵大小,依据灰度熵大小对子块进行分类,针对不同类别子块,设定不同K-奇异值分解算法参数,分别进行字典训练,从而得到多个不同的字典。根据灰度熵大小选择训练好的字典对待表示图像子块进行稀疏表示。仿真实验及结果表明,所提算法能够对图像进行较好的稀疏表示,图像的重构效果也得到了明显提升。

关键词: 稀疏表示, 字典学习, K-奇异值分解, 灰度熵

Abstract: Aiming at the problem that the traditional dictionary learning algorithm of image sparse representation only learns a single dictionary for image training,and can not optimally sparsely represent image blocks containing different image information,through introducing the local gray entropy of image into the dictionary learning algorithm,this paper proposed an adaptive dictionary learning algorithm based on image local gray entropy.The proposed algorithm makes use of the image database as training sample.Firstly,the image database is divided into blocks,and the gray entropy of each sub-block is calculated.Then,the sub-blocks are classified according to the size of the gray entropy,and different K-Singular Value Decomposition (K-SVD) parameters are set for different categories of sub-blocks to perform dictionary training respectively,thus obtaining a plurality of different dictionaries.Lastly,a well-trained dictionary is selected for the image sub-blocks to conduct sparse representation according to the size of the gray entropy.Simulation experiment results show that the proposed algorithm can sparsely represent the images better,and the effect of image reconstruction is also improved significantly.

Key words: Sparse representation, Dictionary learning, K-Singular value decomposition, Gray entropy

中图分类号: 

  • TP391.4
[1]CANDES E J,ROMBERG T,TAO T.Robust uncertainty principles:exact signal reconstruction from highly incomplete frequency information.IEEE Transactions on Information Theo-ry,2006,52(2):489-509.
[2]DONOHO D L.Compressed sensing[J].IEEE Transactions on Inform Theory,2006,52(4):1289-1306.
[3]SHRIVIDYA G,BHARATHI S H.Application of Compressed Sensing on Magnetic Resonance Imaging:A brief survey∥IEEE International Conference on Recent Trends in Electronics, Information & Communication Technology.Bangalore,India,2016:2037-2041.
[4]LIAN Q S,SHI B S,CHEN S Z.Research Advances on Dictio-nary Learning Models,Algorithms and Applications[J].Acta Automatica Sinica,2015,41(2):240-260.(in Chinese)练秋生,石保顺,陈书贞.字典学习模型、算法及其应用研究进展[J].自动化学报,2015,41(2):240-260.
[5]RENE V,YI M,SHANKAR S.Generalized principal compo-nent analysis[J].IEEE Transactions Pattern Anal Mach Intell,2005,27(12):1945-1959.
[6]LIU Z,SONG X N,YU D J,et al.Super-resolution reconstruction algorithm based on multi-component dictionary and sparse representation[J].Journal of Nanjing University of Science and Technology,2014,38(1):1-5.(in Chinese)刘梓,宋晓宁,於东军,等.基于多成分字典和稀疏表示的超分辨率重建算法[J].南京理工大学学报,2014,38(1):1-5.
[7]YANG S Y,JIN H H,WANG M,et al.Data-Driven Compressive Sampling and Learning Sparse Coding for Hyperspectral Image Classification[J].IEEE Geoscience and Remote Sensing Letters,2014,11(2):479-483.
[8]LIU X M,LIU Y M.Color image denoising with block K-SVD dictionary learning[J].Journal of Nanjing University of Science and Technology,2016,40(5):607-612.(in Chinese)刘晓曼,刘永民.基于分块K-SVD字典学习的彩色图像去噪[J].南京理工大学学报,2016,40(5):607-612.
[9]ENGAN K,AASE S O,HUSOY J H.Method of optimal direc-tions for frame design.http://xueshu.baidu.com/usercenter/paper/show?paperid=dc1a54ef946aaaa15e2aa439a9d-116c9&site=xueshu_se.
[10]MAIRAL J,BACH F,PONCE J,et al.Online learning for matrix factorization and sparse coding[J].Journal of Machine Learning Research,2010,11(1):19-60.
[11]AHARON M,ELAD M,BRUCKSTEIN A.K-SVD:an algo-rithm for designing overcomplete dictionaries for sparse representation[J].IEEE Transaction on Signal Processing,2006,54(11):4311-4322.
[12]MITTU G P,VIVEK M,JOONKI P.Imaging inverse problemusing sparse representation with adaptive dictionary learning[C]∥IEEE International Advance Computing Conference.2015.
[13]ZHANG S Y.Research on Image Compressive Sensing Techno-logy Based on Redundant Dictionary[D].Jiling:JiLing University,2016.(in Chinese)张书扬.基于冗余字典的图像压缩感知技术研究[D].吉林:吉林大学,2016.
[14]CONG Y L,ZHANG S Y,LIAN Y Y.K-SVD dictionary lear-ning and image reconstructionbased on variance of image patches[C]∥8th International Symposium on Computational Intelligence and Design.2015:254-257.
[15]SUN J D,ZHAO H H.Sparse Representation and Applications in Image Processing[J].Infrared Technology,2014,36(7):533-537.(in Chinese)孙君顶,赵慧慧.图像稀疏表示及其在图像处理中的应用[J].红外技术,2014,36(7):533-537.
[16]CANDES E,ROMBERG J,TAO T.Stablesignal recovery from incomplete and inaccurate measurements[J].Communications on Pure and Applied Mathematics,2006,59(8):1207-1223.
[17]ZHANG D S,ZHANG L H.Research on fast dictionary learning algorithm under compressed sensing framework[J].Research and exploration in laboratory,2015,34(11):94-98.(in Chinese)张得生,张莉华.压缩感知框架下快速字典的学习算法[J].实验室研究与探索,2015,34(11):94-98.
[18]ZHANG Y L,WANG Y,LU H Z.Block objects detection based on entropy of brightness[J].Systems Engineering and Electro-nic,2008,30(2):201-204.(in Chinese)张永亮,汪洋,卢焕章.基于图像灰度熵的团块目标检测方法[J].系统工程与电子技术,2008,30(2):201-204.
[19]KHANH Q D,HIUK J S,JEON B.Weighted Overlapped Recovery for Blocking Artefacts Reduction in Block-based Compressive Sensing of Images[J].Electronics Letters,2015,51(1):48-50.
[20]ZHANG B,LIU Y L.A novel block compressed sensing based on matrix permutation∥Visual Communications and Image Processing.2016:1-4. [21]ZHU X,LIU L,JIN P.Morphological component decomposition combined with compressed sensing for image compression∥2016 IEEE International Conference on Information and Automation.2016:1726-1731.
[1] 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[J]. 计算机科学, 2019, 46(9): 254-258.
[2] 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[J]. 计算机科学, 2019, 46(7): 217-223.
[3] 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[J]. 计算机科学, 2019, 46(7): 238-245.
[4] 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[J]. 计算机科学, 2019, 46(6A): 188-191.
[5] 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[J]. 计算机科学, 2019, 46(4): 66-72.
[6] 毛峡, 王岚, 李建军. 一种基于RGB-D特征融合的人体行为识别框架[J]. 计算机科学, 2018, 45(8): 22-27.
[7] 甘玲, 赵福超, 杨梦. 一种自适应组稀疏表示的图像修复方法[J]. 计算机科学, 2018, 45(8): 272-276.
[8] 贾旭, 孙福明, 李豪杰, 曹玉东. 基于有监督双正则NMF的静脉识别算法[J]. 计算机科学, 2018, 45(8): 283-287.
[9] 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[J]. 计算机科学, 2018, 45(7): 264-270.
[10] 张玉雪,唐振民,钱彬,徐威. 基于稀疏表示和多特征融合的路面裂缝检测[J]. 计算机科学, 2018, 45(7): 271-277.
[11] 李键红,吴亚榕,吕巨建. 基于组稀疏表示的在线单帧图像超分辨率算法[J]. 计算机科学, 2018, 45(4): 312-318.
[12] 蔡体健,樊晓平,陈志杰,廖志芳. 非共享多测量向量的稀疏表示分类模型[J]. 计算机科学, 2018, 45(3): 258-262.
[13] 蔡柳萍, 解辉, 张福泉, 张龙飞. 基于稀疏表示和特征加权的大数据挖掘方法的研究[J]. 计算机科学, 2018, 45(11): 256-260.
[14] 王威, 陈俊伍, 王新. 自适应加权特征字典与联合稀疏相结合的遥感目标检测[J]. 计算机科学, 2018, 45(10): 276-280.
[15] 季冲,王胜,陆建峰. 基于Fisher鉴别字典学习的人体行为识别[J]. 计算机科学, 2017, 44(7): 270-274.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 雷丽晖,王静. 可能性测度下的LTL模型检测并行化研究[J]. 计算机科学, 2018, 45(4): 71 -75, 88 .
[2] 孙启,金燕,何琨,徐凌轩. 用于求解混合车辆路径问题的混合进化算法[J]. 计算机科学, 2018, 45(4): 76 -82 .
[3] 张佳男,肖鸣宇. 带权混合支配问题的近似算法研究[J]. 计算机科学, 2018, 45(4): 83 -88 .
[4] 钟菲,杨斌. 基于主成分分析网络的车牌检测方法[J]. 计算机科学, 2018, 45(3): 268 -273 .
[5] 王振武,吕小华,韩晓辉. 基于四叉树分割的地形LOD技术综述[J]. 计算机科学, 2018, 45(4): 34 -45 .
[6] 李珊,饶文碧. 基于视频的矿井中人体运动区域检测[J]. 计算机科学, 2018, 45(4): 291 -295 .
[7] 翟鹏珺,方贤文,刘祥伟. 基于接口变迁的交互流程模型挖掘方法[J]. 计算机科学, 2018, 45(3): 317 -321 .
[8] 郭帅,刘亮,秦小麟. 用户偏好约束的空间关键词范围查询处理方法[J]. 计算机科学, 2018, 45(4): 182 -189 .
[9] 戴文静, 袁家斌. 隐含子群问题的研究现状[J]. 计算机科学, 2018, 45(6): 1 -8 .
[10] 张景,朱国宾. 基于CBOW-LDA主题模型的Stack Overflow编程网站热点主题发现研究[J]. 计算机科学, 2018, 45(4): 208 -214 .