计算机科学 ›› 2018, Vol. 45 ›› Issue (8): 283-287.doi: 10.11896/j.issn.1002-137X.2018.08.051

• 图形图像与模式识别 • 上一篇    下一篇

基于有监督双正则NMF的静脉识别算法

贾旭1, 孙福明1, 李豪杰2, 曹玉东1   

  1. 辽宁工业大学电子与信息工程学院 辽宁 锦州 1210011
    大连理工大学软件学院 辽宁 大连1160242
  • 收稿日期:2017-06-16 出版日期:2018-08-29 发布日期:2018-08-29
  • 作者简介:贾 旭(1983-),男,博士,副教授,CCF会员,主要研究方向为模式识别、机器学习,E-mail:gbjdjiaxu@163.com(通信作者); 孙福明(1972-),男,博士,教授,主要研究方向为多媒体处理、机器学习; 李豪杰(1973-),男,博士,教授,主要研究方向为多媒体信息检索、计算机视觉; 曹玉东(1971-),男,博士,副教授,主要研究方向为图像处理。
  • 基金资助:
    本文受国家自然科学基金(61502216,61572244)资助。

Vein Recognition Algorithm Based on Supervised NMF with Two Regularization Terms

JIA Xu1, SUN Fu-ming1, LI Hao-jie2, CAO Yu-dong1   

  1. School of Electronics & Information Engineering,Liaoning University of Technology,Jinzhou,Liaoning 121001,China1
    School of Software,Dalian University of Technology,Dalian,Liaoning 116024,China2
  • Received:2017-06-16 Online:2018-08-29 Published:2018-08-29

摘要: 为使提取的静脉图像特征具有较好的聚类特性以更利于正确识别,提出了一种基于有监督非负矩阵分解的识别算法。首先,对静脉图像进行分块处理,通过融合所有的子图像特征形成静脉的原始特征;其次,采用特征的稀疏性与聚类属性双正则项,对原始的非负矩阵分解模型进行改进;然后,基于梯度下降法对改进的非负矩阵分解模型进行求解,实现对原始特征的降维与优化;最后,利用最近邻算法对新的特征进行匹配,从而获得识别结果。实验结果表明,对于3种静脉样本数据库,所提识别算法的错误接受率与错误拒绝率分别可以达到0.02与0.03;此外,其2.89s的识别时间可以满足实时性要求。

关键词: 静脉识别, 生物特征, 非负矩阵分解, 特征降维, 稀疏表示

Abstract: In order to make the extracted vein feature have good clustering performance and thus be more conductive to correct identification,this paper proposed a recognition algorithm based on supervised Nonnegative Matrix Factorization (NMF).Firstly,vein image is divided into blocks,and the original vein feature can be acquired by fusing all sub image features.Secondly,the sparsity and clustering property of feature vectors areregarded as two regularization terms,and the original NMF model is improved.Then,gradient descent method is used to solve the improved NMF model,and feature optimization and dimension reduction can be achieved.Finally,by using nearest neighbor algorithm to match new vein features,the recognition results can be acquired.Experiment results show that the obtained false accept rate (FAR) and false reject rate (FRR) of the proposed recognition algorithm can be reached 0.02 and 0.03 respectively for three vein databases,in addition,the recognition time of 2.89 seconds can meet real-time requirement.

Key words: Vein recognition, Biological feature, Nonnegative Matrix Factorization, Feature dimension reduction, Sparse representation

中图分类号: 

  • TP391.4
[1]JIA X,SUN F M,CAO Y D,et al.Dorsal hand vein recognition algorithm based on effective dimensional feature [J].Computer Science,2016,43(1):315-318.(in Chinese)贾旭,孙福明,曹玉东,等.基于有效维度特征的手背静脉识别算法[J].计算机科学,2016,43(1):315-318.
[2]WANG K J,YUAN Z.Finger vein recognition based on wavelet moment fused with PCA transform [J].Pattern Recognition & Artificial Intelligence,2007,20(5):692-697.(in Chinese)王科俊,袁智.基于小波矩融合PCA变换的手指静脉识别[J].模式识别与人工智能,2007,20(5):692-697.
[3]LI T G,MA S L,ZHANG Z B,et al.Hand vein recognitionmethod based on Bandelet transformation [J].Journal of Jilin University (Science Edition),2007,45(6):975-978.(in Chinese)李铁钢,马驷良,张忠波,等.基于Bandelet变换的手背静脉识别算法[J].吉林大学学报(理学版),2007,45(6):975-978.
[4]HAN W Y,LEE J C.Palm vein recognition using adaptive Gabor filter [J].Expert Systems with Applications,2012,39(18):13225-13234.
[5]HSU C B,LEE J C,HAO S S,et al.Dorsal hand vein recognition using Gabor feature-based 2-directional principal component analysis[J].Advanced Science Letters,2012,8(1):813-817.
[6]WANG R,WANG G Y,CHEN Z,et al.A palm vein identification system based on Gabor wavelet features [J].Neural Computing and Applications,2014,24(1):161-168.
[7]ALI M A J,BU W,WU X Q,et al.Palm vein verification using Gabor Filter [J].International Journal of Computer Science Issues,2013,10(1):678-684.
[8]SUN J W,ABDULLA W.Palm vein recognition by combining Curvelet transform and Gabor filter[C]∥Biometric Recognition 8th Chinese Conference.Jinan,China,2013:314-321.
[9]JIA X,XUE D Y,CUI J J,et al.Dorsal hand vein recognition algorithm based on Ridgelet transformation of divided blocks [J].Pattern Recognition & Artificial Intelligence,2011,24(3):346-352.(in Chinese)贾旭,薛定宇,崔建江,等.基于分块脊波变换的手背静脉识别算法[J].模式识别与人工智能,2011,24(3):346-352.
[10]HSU C B,LEE J C,CHUANG S J,et al.Gaussian directional pattern for dorsal hand vein recognition[J].The Imaging Science Journal,2014,63(1):54-62.
[11]LEE J C,LO T M,CHANG C P.Dorsal hand vein recognition based on directional filter bank[J].Signal,Image and Video Processing,2016,10(1):145-152.
[12]PREMALATHA K,ANANTHA T,NATARAJAN M.A Dorsal Hand Vein Recognition based on Local Gabor Phase Quantization with Whitening Transformation[J].Defence Science Journal,2014,64(2):159-167.
[13]KUMAR A,PRATHYUSHA K V.Personal authenticationusing hand vein triangulation and knuckle shape[J].IEEE Transactions on Image Processing,2009,9(18):2127-2136.
[14]HU Y P,WANG Z Y,YANG X P,et al.Hand vein recognition based on the connection lines of reference point and feature point [J].Infrared Physics & Technology,2014,62(1),110-114.
[15]WANG Y D,LIAO W P.Hand vein recognition based on feature coding[C]∥Proceedings of the 7th Chinese conference on Biometric Recognition.Guangzhou,China,2012,165-175.
[16]HUANG D,TANG Y,WANG Y,et al.Hand-Dorsa Vein Reco-gnition by Matching Local Features of Multisource Keypoints[J].IEEE Transactions on Cybernetics,2015,45(9):1823-1837.
[17]HUANG D,ZHU X,WANG Y,et al.Dorsal hand vein recognition via hierarchical combination of texture and shape clues[J].Neurocomputing,2016,214(C):815-828.
[18]ANIKA P,DANIEL H,CHRISTOPH B.Feature extractionfrom vein images using spatial information and chain codes[J].Information Security Technical Report,2012,17(1/2):26-35.
[19]XI X,YANG L,YIN Y.Learning discriminative binary codes for finger vein recognition[J].Pattern Recognition,2017,66:26-33.
[20]LEE D D,SEUNG H S.Learning the parts of objects by non-negative matrix factorization[J].Nature,1999,401(6755):788-791.
[21]ZHANG Z,XU Y,YANG J,et al.A Survey of Sparse Representation:Algorithms and Applications [J].IEEE Access,2015,3:490-530.
[22]YIN Y L,LIU L L,SUN X W.SDUMLA-HMT:A Multimodal Biometric Database[C]∥Chinese Conference of BiometricRecog-nition.Beijing,China,2011:260-268.
[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] 康林瑶, 唐兵, 夏艳敏, 张黎. 基于GPU加速和非负矩阵分解的并行协同过滤推荐算法[J]. 计算机科学, 2019, 46(8): 106-110.
[3] 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.
[4] 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.
[5] 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.
[6] 何孝文, 胡一飞, 王海平, 陈默. 在线学习非负矩阵分解[J]. 计算机科学, 2019, 46(6A): 473-477.
[7] 黄梦婷, 张灵, 姜文超. 基于流形正则化的多类型关系数据联合聚类方法[J]. 计算机科学, 2019, 46(6): 64-68.
[8] 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[J]. 计算机科学, 2019, 46(5): 266-271.
[9] 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.
[10] 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.
[11] 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.
[12] 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.
[13] 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.
[14] 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.
[15] 于晓,聂秀山,马林元,尹义龙. 基于短空时变化的鲁棒视频哈希算法[J]. 计算机科学, 2018, 45(2): 84-89.
Viewed
Full text


Abstract

Cited

  Shared   
  Discussed   
[1] 编辑部. 新网站开通,欢迎大家订阅![J]. 计算机科学, 2018, 1(1): 1 .
[2] 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 .
[3] 夏庆勋,庄毅. 一种基于局部性原理的远程验证机制[J]. 计算机科学, 2018, 45(4): 148 -151, 162 .
[4] 厉柏伸,李领治,孙涌,朱艳琴. 基于伪梯度提升决策树的内网防御算法[J]. 计算机科学, 2018, 45(4): 157 -162 .
[5] 王欢,张云峰,张艳. 一种基于CFDs规则的修复序列快速判定方法[J]. 计算机科学, 2018, 45(3): 311 -316 .
[6] 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 .
[7] 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 .
[8] 伍建辉,黄中祥,李武,吴健辉,彭鑫,张生. 城市道路建设时序决策的鲁棒优化[J]. 计算机科学, 2018, 45(4): 89 -93 .
[9] 刘琴. 计算机取证过程中基于约束的数据质量问题研究[J]. 计算机科学, 2018, 45(4): 169 -172 .
[10] 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 .