基于迁移孪生非负矩阵分解的静脉识别算法

doi:10.11772/j.issn.1001-9081.2020060965

计算机应用 ›› 2021, Vol. 41 ›› Issue (3): 898-903.DOI: 10.11772/j.issn.1001-9081.2020060965

所属专题：前沿与综合应用

基于迁移孪生非负矩阵分解的静脉识别算法

王锦凯, 贾旭

辽宁工业大学电子与信息工程学院, 辽宁锦州 121001

收稿日期:2020-07-06 修回日期:2020-10-21 出版日期:2021-03-10 发布日期:2021-01-15
通讯作者: 贾旭
作者简介:王锦凯(1982-),男,辽宁锦州人,实验师,硕士,主要研究方向:计算机视觉、机器学习;贾旭(1983-),男,辽宁开原人,副教授,博士,CCF会员,主要研究方向:模式识别、机器学习。
基金资助:
国家自然科学基金资助项目（61502216）；辽宁省自然科学基金指导计划项目（2019-ZD-0700）。

Vein recognition algorithm based on Siamese nonnegative matrix factorization with transferability

WANG Jinkai, JIA Xu

School of Electronics and Information Engineering, Liaoning University of Technology, Jinzhou Liaoning 121001, China

Received:2020-07-06 Revised:2020-10-21 Online:2021-03-10 Published:2021-01-15
Supported by:
This work is partially supported by the National Natural Science Foundation of China (61502216), the Guidance Plan of Liaoning Natural Science Foundation (2019-ZD-0700).

摘要/Abstract

摘要： 针对某一静脉图像数据集下获得的识别算法对于其他数据集缺少普适性的问题，提出了一种具有迁移性的孪生非负矩阵分解（NMF）模型。首先，通过采用两个结构相同且参数共享的NMF模型实现了对源数据集中带有相同标签静脉图像的有监督学习；然后，通过使用最大均值差异约束降低了不同数据集之间静脉特征的差异性，即将源数据集中的知识迁移至目标数据集中；最后，基于余弦距离实现静脉图像的匹配。实验结果表明，所提的识别算法不仅可以在源数据集下上获得较高的正确识别率，而且仅利用目标数据集中的少量静脉图像便可使得在目标数据集上的平均错误接受率（FAR）与平均错误拒绝率（FRR）分别降低至0.043与0.055。此外，所提算法平均0.56 s的识别时间可以满足识别的实时性要求。

关键词: 静脉识别, 孪生模型, 非负矩阵分解, 迁移学习, 梯度下降法

Abstract: Concerning the problem that the recognition algorithm which was obtained under one vein image dataset is lack of universality to other datasets, a Siamese Nonnegative Matrix Factorization (NMF) model with transferability was proposed. Firstly, the supervised learning for the vein images with same labels in the source dataset was achieved by using two NMF models with the same structures and the parameter sharing. Then, the vein feature differences between two different datasets were reduced through using maximum mean discrepancy constraint, that is to transfer the knowledge in the source dataset to the target dataset. Finally, the matching of vein images was realized based on cosine distance. Experimental results show that, the proposed recognition algorithm can not only achieve the high recognition accuracy on the source dataset, but also respectively reduce the average False Accept Rate (FAR) and average False Reject Rate (FRR) to 0.043 and 0.055 on the target dataset when using only a small number of vein images in the target dataset. In addition, the average recognition time of the proposed algorithm is 0.56 seconds, which can meet the real-time requirement of recognition.

Key words: vein recognition, Siamese model, Nonnegative Matrix Factorization (NMF), transfer learning, gradient descent algorithm

中图分类号:

TP391.4

王锦凯, 贾旭. 基于迁移孪生非负矩阵分解的静脉识别算法[J]. 计算机应用, 2021, 41(3): 898-903.

WANG Jinkai, JIA Xu. Vein recognition algorithm based on Siamese nonnegative matrix factorization with transferability[J]. Journal of Computer Applications, 2021, 41(3): 898-903.

参考文献

[1] 贾旭, 王锦凯, 崔建江, 等. 基于稀疏编码的手背静脉识别算法[J]. 计算机应用,2015,35(4):1129-1132, 1153.(JIA X, WANG J K,CUI J J,et al. Dorsal hand vein recognition algorithm based on sparse coding[J]. Journal of Computer Applications, 2015,35(4):1129-1132,1153.)
[2] MALUTAN R,EMERICH S,CRISAN S,et al. Dorsal hand vein recognition based on Riesz wavelet transform and local line binary pattern[C]//Proceedings of the 3rd International Conference on Frontiers of Signal Processing. Piscataway:IEEE,2017:146-150.
[3] 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.(LI T G,MA S L,ZHANG Z B,et al. Hand vein recognition method based on Bandelet transformation[J]. Journal of Jilin University (Science Edition),2007,45(6):975-978.)
[4] SUN J,ABDULLA W. Palm vein recognition by combining curvelet transform and Gabor filter[C]//Proceedings of the 2013 Chinese Conference on Biometric Recognition, LNCS 8232. Cham:Springer,2013:314-321.
[5] 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.
[6] WANG R,WANG G,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] MA X,JING X,HUANG H,et al. Palm vein recognition scheme based on an adaptive Gabor filter[J]. IET Biometrics,2017,6(5):325-333.
[8] HSU C B,LEE J C,CHUANG S J,et al. Gaussian directional pattern for dorsal hand vein recognition[J]. The Imaging Science Journal,2015,63(1):54-62.
[9] LEE J C,LEE C H,HSU C B,et al. Dorsal hand vein recognition based on 2D Gabor filters[J]. The Imaging Science Journal,2014, 62(3):127-138.
[10] LADOUX P O,ROSENBERGER C,DORIZZI B. Palm vein verification system based on SIFT matching[C]//Proceedings of the 2009 International Conference on Biometrics,LNCS 5558. Berlin:Springer,2009:1290-1298.
[11] 李秀艳, 刘铁根, 邓仕超, 等. 基于SURF算子的快速手背静脉识别[J]. 仪器仪表学报,2011,32(4):831-836.(LI X Y,LIU T G,DENG S C,et al. Fast recognition of hand vein with SURF descriptors[J]. Chinese Journal of Scientific Instrument,2011,32(4):831-836.)
[12] YANG W,WANG S,HU J,et al. A fingerprint and finger-vein based cancelable multi-biometric system[J]. Pattern Recognition, 2018,78:242-251.
[13] JIA X,SUN F,LI H,et al. Hand vein recognition algorithm based on NMF with sparsity and clustering property constraints in feature mapping space[J]. Chinese Journal of Electronics,2019, 28(6):1184-1190.
[14] KUZU R S,PICIUCCO E,MAIORANA E,et al. On-the-fly finger-vein-based biometric recognition using deep neural networks[J]. IEEE Transactions on Information Forensics and Security, 2020,15:2641-2654.
[15] KIM W, SONG J M, PARK K R. Multimodal biometric recognition based on convolutional neural network by the fusion of finger-vein and finger shape using Near-InfRared(NIR) camera sensor[J]. Sensors,2018,18(7):No. 2296.
[16] XIE C,KUMAR A. Finger vein identification using convolutional neural network and supervised discrete hashing[J]. Pattern Recognition Letters,2019,119:148-156.
[17] KANG W,LU Y,LI D,et al. From noise to feature:exploiting intensity distribution as a novel soft biometric trait for finger vein recognition[J]. IEEE Transactions on Information Forensics and Security,2019,14(4):858-869.
[18] SHAZEEDA S,ROSDI B A. Finger vein recognition using mutual sparse representation classification[J]. IET Biometrics,2019,8(1):49-58.
[19] KUMAR A,PRATHYUSHA K V. Personal authentication using hand vein triangulation and knuckle shape[J]. IEEE Transactions on Image Processing,2009,18(9):2127-2136.
[20] HU Y,WANG Z,YANG X,et al. Hand vein recognition based on the connection lines of reference point and feature point[J]. Infrared Physics and Technology,2014,62:110-114.
[21] XI X,YANG L,YIN Y. Learning discriminative binary codes for finger vein recognition[J]. Pattern Recognition, 2017, 66:26-33.
[22] WANG G,SUN C,SOWMYA A. Multi-weighted co-occurrence descriptor encoding for vein recognition[J]. IEEE Transactions on Information Forensics and Security,2019,15:375-390.
[23] LIU H,YANG G,YANG L,et al. Learning personalized binary codes for finger vein recognition[J]. Neurocomputing,2019, 365:62-70.
[24] YANG L,YANG G,XI X,et al. Finger vein code:from indexing to matching[J]. IEEE Transactions on Information Forensics and Security,2019,14(5):1210-1223.
[25] SU K,YANG G,YANG L,et al. Learning binary hash codes for finger vein image retrieval[J]. Pattern Recognition Letters,2019, 117:74-82.
[26] HE A,LUO C,TIAN X,et al. A twofold Siamese network for real-time object tracking[C]//Proceedings of the 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Piscataway:IEEE,2018:4834-4843.
[27] LONG M,ZHU H,WANG J,et al. Deep transfer learning with joint adaptation networks[C]//Proceedings of the 34th International Conference on Machine Learning. New York:JMLR. org,2017:2208-2217.
[28] WU W, KWONG S, ZHOU Y, et al. Nonnegative matrix factorization with mixed hypergraph regularization for community detection[J]. Information Sciences,2018,435:263-281.
[29] LIU J,ZHANG Y. Palm-dorsa vein recognition based on twodimensional fisher linear discriminant[C]//Proceedings of the 2011 International Conference on Image Analysis and Signal Processing. Piscataway:IEEE,2011:550-552.
[30] YIN Y,LIU L,SUN X. SDUMLA-HMT:a multimodal biometric database[C]//Proceedings of the 2011 Chinese Conference on Biometric Recognition,LNCS 7098. Berlin:Springer,2011:260-268.

基于迁移孪生非负矩阵分解的静脉识别算法

Vein recognition algorithm based on Siamese nonnegative matrix factorization with transferability

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]	陈争涛, 黄灿, 杨波, 赵立, 廖勇. 基于迁移学习的并行卷积神经网络牦牛脸识别算法[J]. 计算机应用, 2021, 41(5): 1332-1336.
[2]	黄学雨, 徐浩特, 陶剑文. 具有特征选择的多源自适应分类框架[J]. 计算机应用, 2020, 40(9): 2499-2506.
[3]	张晨曦, 唐曙, 唐珂. 迁移学习下的火箭发动机参数异常检测策略[J]. 计算机应用, 2020, 40(9): 2774-2780.
[4]	白茹意, 郭小英, 贾春花. 基于特征融合的小样本抽象画图像情感预测[J]. 计算机应用, 2020, 40(8): 2207-2213.
[5]	马金林, 魏萌, 马自萍. 基于深度迁移学习的肺结节分割方法[J]. 计算机应用, 2020, 40(7): 2117-2125.
[6]	肖禾, 刘志勤, 王庆凤, 黄俊, 周莹, 刘启榆, 徐卫云. 基于迁移学习的多视角乳腺肿块和钙化簇分类方法[J]. 计算机应用, 2020, 40(5): 1460-1464.
[7]	谌贵辉, 易欣, 李忠兵, 钱济人, 陈伍. 基于改进YOLOv2和迁移学习的管道巡检航拍图像第三方施工目标检测[J]. 计算机应用, 2020, 40(4): 1062-1068.
[8]	王锦凯, 贾旭. 基于加权正交约束非负矩阵分解的车脸识别算法[J]. 计算机应用, 2020, 40(4): 1050-1055.
[9]	陈杰, 张挺, 杜奕. 基于自适应深度迁移学习的多孔介质重构[J]. 计算机应用, 2020, 40(4): 1231-1236.
[10]	娄梦莹, 王天景, 刘娅琴, 杨丰, 黄靖. 基于侧链连接卷积神经网络的手掌静脉图像识别[J]. 计算机应用, 2020, 40(12): 3673-3678.
[11]	成其伟, 陈启买, 贺超波, 刘海. 基于改进对称二值非负矩阵分解的重叠社区发现方法[J]. 计算机应用, 2020, 40(11): 3203-3210.
[12]	郑宗生, 胡晨雨, 姜晓轶. 基于改进的最大均值差异算法的深度迁移适配网络[J]. 计算机应用, 2020, 40(11): 3107-3112.
[13]	毛文涛, 杨超, 刘亚敏, 田思雨. 面向轴承早期故障检测的多尺度残差注意力深度领域适配模型[J]. 计算机应用, 2020, 40(10): 2890-2898.
[14]	武宽, 秦品乐, 柴锐, 曾建朝. 基于不同超声成像的甲状腺结节良恶性判别[J]. 计算机应用, 2020, 40(1): 77-82.
[15]	李旻择, 李小霞, 王学渊, 孙维. 基于多尺度核特征卷积神经网络的实时人脸表情识别[J]. 计算机应用, 2019, 39(9): 2568-2574.