基于深度卷积神经网络的指纹纹型分类算法

doi:10.7523/j.issn.2095-6134.2016.06.013

中国科学院大学学报 ›› 2016, Vol. 33 ›› Issue (6): 808-814.DOI: 10.7523/j.issn.2095-6134.2016.06.013

基于深度卷积神经网络的指纹纹型分类算法

江璐¹, 赵彤^2,3, 吴敏²

1. 中国科学院大学计算机与控制学院, 北京 101408;
2. 中国科学院大学数学科学学院, 北京 100049;
3. 中国科学院大数据挖掘与知识管理重点实验室, 北京 100049

收稿日期:2016-03-16 发布日期:2016-11-15
通讯作者: 赵彤,E-mail:zhaotong@ucas.ac.cn
基金资助:
国家自然基金（11331012，11571014，71271204）资助

Fingerprint pattern classification algorithm based on deep convolutional neural networks

JIANG Lu¹, ZHAO Tong^2,3, WU Min²

1. School of Computer and Control, University of Chinese Academy of Sciences, Beijing 101408, China;
2. School of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing 100049, China;
3. Key Laboratory of Big Data Mining and Knowledge Management, Chinese Academy of Sciences, Beijing 100049, China

Received:2016-03-16 Published:2016-11-15

摘要/Abstract

摘要：

传统指纹纹型分类算法的准确率直接受到相应特征提取算法的影响.在海量指纹库中，同类纹型指纹形态变化明显增大，不同类纹型界限变得模糊，仅通过人工定义的特征进行分类很难适应全部指纹数据.为解除纹型分类问题与人工定义的特征提取问题的耦合，提出一种直接在指纹原图上进行纹型识别的算法.利用卷积神经网络自动特征提取的能力从大量指纹数据中学习得到纹型特征，并通过对训练数据的设计使网络能够适应指纹的多样性，提升算法的鲁棒性.此外，多尺度网络模型平均方法使分类准确性得到进一步提升.在国际公开指纹数据集NIST DB4上测得纹型四分类准确率达94.2%.

关键词: 指纹纹型分类, 卷积神经网络, 指纹识别

Abstract:

The accuracies of traditional fingerprint pattern classification algorithms rely heavily on the corresponding feature extraction algorithms. Further more, the within-class variance of fingerprint patterns increases while the between-class variance decreases in large-scale database. So it is difficult for hand-designed features to suit with all fingerprint data. In order to remove the coupling with hand-designed feature extraction algorithms, we propose an approach to directly recognize patterns in raw fingerprint images. It takes advantage of the automatic feature extraction ability of convolutional neural networks to learn patterns from large amount of images. The training data are carefully designed to fit the variety of fingerprints and to improve the robutness. Meanwhile, the accuracy is further improved by averaging multi-scale models. In our experiment, an accuracy of 94.2% for four-class classification has been achieved in the international opening fingerprint dataset NIST DB 4. Our algorithm surpasses many classical algorithms, and it is both practical and meaningful.

Key words: fingerprint pattern classification, convolutional neural networks, fingerprint recognition

中图分类号:

TP391

江璐, 赵彤, 吴敏. 基于深度卷积神经网络的指纹纹型分类算法[J]. 中国科学院大学学报, 2016, 33(6): 808-814.

JIANG Lu, ZHAO Tong, WU Min. Fingerprint pattern classification algorithm based on deep convolutional neural networks[J]. , 2016, 33(6): 808-814.

参考文献

[1] Henry E R. Classification and uses of finger prints[M].3rd ed. London:HM Stationery Office, 1905.
[2] Wilson C L, Candela G T, Watson C I. Neural network fingerprint classification[J]. Journal of Artificial Neural Networks, 1994, 1(2):203-228.
[3] Yao Y, Frasconi P, Pontil M. Fingerprint classi-fication with combinations of support vector ma-chines[C]//Audio and Video Based Biometric Person Authentication. Springer Berlin Heidelberg, 2001:253-258.
[4] Cao K, Pang L, Liang J, et al. Fingerprint classifica-tion by a hierarchical classifier[J]. Pattern Recognition, 2013, 46(12):3186-3197.
[5] Jung H W, Lee J H. Noisy and incomplete fingerprint classification using local ridge distribution models[J]. Pattern Recognition, 2015, 48(2):473-484.
[6] Liu M. Fingerprint classification based on singulari-ties[C]//Pattern Recognition, 2009. Chinese Conference on. IEEE, 2009:1-5.
[7] Leung K C, Leung C H. Improvement of fingerprint re-trieval by a statistical classifier[J]. Information Foren-sics and Security, IEEE Transactions on, 2011, 6(1):59-69.
[8] Luo J, Song D, Xiu C, et al. Fingerprint classification combining curvelet transform and gray-level cooccur-rence matrix[J]. Mathematical Problems in Engineering, 2014(7):1-15.
[9] Bouvrie J. Notes on convolutional neural networks[J/OL].Neural Nets, 2006.[2016-02-25].http://people.csail.mit.edu/jvb/papers/cnn.tutorial_old.pdf.
[10] Ciresan D C, Meier U, Gambardella L M, et al. Deep, big, simple neural nets for handwritten digit recogni-tion[J]. Neural Computation, 2010, 22(12):3207-3220.
[11] Ciresan D C, Meier U, Gambardella L M, et al.Convolutional neural network committees for hand-written character classification[C]//Proceedings of the 2011 International Conference on Document Analysis and Recognition. IEEE Computer Society, 2011:1135-1139.
[12] Lawrence S, Giles C L, Tsoi A C, et al. Face recogni-tion:a convolutional neural-network approach[J]. Neu-ral Networks, IEEE Transactions on, 1997, 8(1):98-113.
[13] Ji S, Xu W, Yang M, et al. 3D convolutional neural net-works for human action recognition[J]. Pattern Anal-ysis and Machine Intelligence, IEEE Transactions on, 2013, 35(1):221-231.
[14] Frassetto Nogueira R, de Alencar Lotufo R, Campos Machado R. Evaluating software-based fingerprint live-ness detection using convolutional networks and local binary patterns[C]//Biometric Measurements and Sys-tems for Security and Medical Applications (BIOM-S) Proceedings, 2014 IEEE Workshop on. IEEE, 2014:22-29.
[15] Cao K, Jain A K. Latent orientation field estima-tion via convolutional neural network[C]//International Conference on Biometrics. IEEE, 2015:349-356.
[16] Watson C I, Wilson C L. NIST special database 4.Fingerprint Database, National Institute of Standards and Technology, 1992, 17:77.
[17] Nair V, Hinton G E. Rectified linear units improve restricted boltzmann machines[C]//Proceedings of the 27th International Conference on Machine Learning (ICML-10). 2010:807-814.
[18] Ciresan D, Giusti A, Gambardella L M, et al. Deep neural networks segment neuronal membranes in elec-tron microscopy images[C]//Advances in Neural Infor-mation Processing Systems. 2012:2843-2851.
[19] Domingos P. A few useful things to know about machine learning[J]. Communications of the ACM, 2012, 55(10):78-87.
[20] Jia Y, Shelhamer E, Donahue J, et al. Caffe:convolutional architecture for fast feature embed-ding[C]//Proceedings of the ACM International Confer-ence on Multimedia. ACM, 2014:675-678.
[21] Karu K, Jain A K. Fingerprint classification[J]. Pattern Recognition, 1996, 29(3):389-404.
[22] Jain A K, Minut S. Hierarchical kernel fitting for fingerprint classification and alignment[C]//Pattern Recognition, 2002. Proceedings 16th International Conference on. IEEE, 2002, 2:469-473.
[23] Hong J H, Min J K, Cho U K, et al. Fingerprint clas-sification using one-vs-all support vector machines dy-namically ordered with naive bayes classifiers[J]. Pat-tern Recognition, 2008, 41(2):662-671.

基于深度卷积神经网络的指纹纹型分类算法

Fingerprint pattern classification algorithm based on deep convolutional neural networks

PDF (PC)

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 9

编辑推荐

Metrics

本文评价

访问统计

联系我们

[1]	胡莉娜, 高盛华. 基于层次化标签的人体解析[J]. 中国科学院大学学报, 2020, 37(6): 820-827.
[2]	冯齐心, 杨辽, 王伟胜, 陈桃, 黄双燕. 基于时序光谱重构的卷积神经网络遥感农作物分类[J]. 中国科学院大学学报, 2020, 37(5): 619-628.
[3]	宋晗, 杨炜暾, 耿修瑞, 赵永超. 基于卷积神经网络与主动学习的高光谱图像分类[J]. 中国科学院大学学报, 2020, 37(2): 169-176.
[4]	王惠君, 赵桂萍, 李良, 张威, 齐荣, 刘珺. 基于卷积神经网络(CNN)的泥质烃源岩TOC预测模型——以鄂尔多斯盆地杭锦旗地区为例[J]. 中国科学院大学学报, 2020, 37(1): 103-112.
[5]	朱思捷, 雷斌, 吴一戎. 基于稠密卷积神经网络的遥感图像自动色彩校正[J]. 中国科学院大学学报, 2019, 36(1): 93-100.
[6]	王春峰, 苏荔, 黄庆明. 基于卷积神经网络的时空融合的无参考视频质量评价方法[J]. 中国科学院大学学报, 2018, 35(4): 544-549.
[7]	李松, 魏中浩, 张冰尘, 洪文. 深度卷积神经网络在迁移学习模式下的SAR目标识别[J]. 中国科学院大学学报, 2018, 35(1): 75-83.
[8]	闫淑筠, 王文杰, 张玉清. 一种有效的Web指纹识别方法[J]. 中国科学院大学学报, 2016, 33(5): 679-685.
[9]	曹来成, 赵建军, 崔翔, 李可. 基于余弦测度下K-means的网络空间终端设备识别[J]. 中国科学院大学学报, 2016, 33(4): 562-569.