Study for lightweight finger vein recognition based on a small sample.
Journal
Scientific reports
ISSN: 2045-2322
Titre abrégé: Sci Rep
Pays: England
ID NLM: 101563288
Informations de publication
Date de publication:
25 May 2024
25 May 2024
Historique:
received:
15
11
2023
accepted:
23
05
2024
medline:
26
5
2024
pubmed:
26
5
2024
entrez:
25
5
2024
Statut:
epublish
Résumé
To address several common problems of finger vein recognition, a lightweight finger vein recognition algorithm by means of a small sample has been proposed in this study. First of all, a Gabor filter is applied to deal with the images for the purpose of that these processed images can simulate a kind of situation of finger vein at low temperature, such that the generalization ability of the algorithm model can be improved as well. By cutting down the amount of convolutional layers and fully connected layers in VGG-19, a lightweight network can be given. Meanwhile, the activation function of some convolutional layers is replaced to protect the network weight that can be updated successfully. After then, a multi-attention mechanism is introduced to the modified network architecture to result in improving the ability of extracting important features. Finally, a strategy based on transfer learning has been used to reduce the training time in the model training phase. Honestly, it is obvious that the proposed finger vein recognition algorithm has a good performance in recognition accuracy, robustness and speed. The experimental results show that the recognition accuracy can arrive at about 98.45%, which has had better performance in comparison with some existing algorithms.
Identifiants
pubmed: 38796559
doi: 10.1038/s41598-024-63002-1
pii: 10.1038/s41598-024-63002-1
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
12002Subventions
Organisme : Public Welfare Technology Application Research Project of Zhejiang Province
ID : LGF22F030005
Informations de copyright
© 2024. The Author(s).
Références
Li, J. & Qu, Z. A survey of application of deep learning in finger vein recognition. J. Front. Comput. Sci. Technol. 17(11), 2557–2579 (2023).
Yang, J. & Shi, Y. Finger–vein ROI localization and vein ridge enhancement. Pattern Recogn. Lett. 33(12), 1569–1579 (2012).
doi: 10.1016/j.patrec.2012.04.018
Yin, Y., Yang, G. & Yang, L. Survey of finger vein recognition study. J. Data Acquis. Process. 30(5), 933–939 (2015).
Liu, W., Lu, H. & Li, Y. Survey on finger vein reconition research. Comput. Sci. 49(S1), 1–11 (2022).
Zhang, J., Lu, Z. & Li, M. GAN-based image augmentation for finger-vein biometric recognition. IEEE Access 7, 183118–183132 (2019).
doi: 10.1109/ACCESS.2019.2960411
Rosdi, B. A., Shing, C. W. & Suandi, S. A. Finger vein recognition using local line binary pattern. Sensors 11(12), 11357–11371 (2011).
doi: 10.3390/s111211357
pubmed: 22247670
pmcid: 3251987
Lee, E., Jung, H. & Kim, D. New finger biometric method using near infrared imaging. Sensors 11(3), 2319–2333 (2011).
doi: 10.3390/s110302319
pubmed: 22163741
pmcid: 3231585
Marios, V. & Evangelos, D. Finger vein segmentation from infrared images based on a modified separable mumford shah model and local entropy thresholding. Comput. Math. Method. Med. 2015, 1–20 (2015).
Ma, H., Hu, N. & Fang, C. The biometric recognition system based on near-infrared finger vein image. Infrared. Phys. Technol. 116, 103734 (2021).
doi: 10.1016/j.infrared.2021.103734
Shakil, S., Arora, D. & Zaidi, T. An optimal method for identification of finger vein using supervised learning. Meas. Sens. 25, 100583 (2023).
doi: 10.1016/j.measen.2022.100583
Yang, L., Zhang, H. & Fang, C. Finger vein image segmentation based on Hessian matrix and LBF model. J. Hefei Normal Univ. 41(3), 24–28+34 (2023).
Qiu, S., Liu, Y. & Zhou, Y. Finger-vein recognition based on dual-sliding window localization and pseudo-elliptical transformer. Expert. Syst. Appl. 64, 618–632 (2016).
doi: 10.1016/j.eswa.2016.08.031
Shaheed, K., Mao, A. & Qureshi, I. Finger-vein presentation attack detection using depthwise separable convolution neural network. Expert. Syst. Appl. 198, 116786 (2022).
doi: 10.1016/j.eswa.2022.116786
Zhou, W., Wang, Y. & Ji, S. Bimodal biological recognition method based on feature fusion. J. Tianjin Univ. Sci. 37(4), 44–48+54 (2022).
Tao, Z., Feng, Y. & Lin, S. Score level fusion of finger vein and finger knuckle print identification based on transfer learning. Comput. Appl. Softw. 36(12), 162–168+183 (2019).
Yang, W., Hui, C. & Chen, Z. FV-GAN: Finger vein representation using generative adversarial networks. IEEE T. Inf. Foren. Sec. 14(9), 2512–2524 (2019).
doi: 10.1109/TIFS.2019.2902819
Choi, J., Noh, K. & Cho, S. Modified conditional generative adversarial network-based optical blur restoration for finger-vein recognition. IEEE Access 8, 16218–16301 (2020).
Karen, S. & Andrew, Z. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014).
He, K., Sun, J. & Tang, X. Guided image filtering. IEEE T. Pattern. Anal. 35(6), 1397–1409 (2013).
doi: 10.1109/TPAMI.2012.213
Huang, J., Tu, M. & Yang, W. Joint attention network for finger vein authentication. IEEE T. Instrum. Meas. 70, 1–11 (2021).
Yang, J., Yang, J. & Shi,Y. Finger-vein segmentation based on multi-channel even-symmetric Gabor filters. In Proceedings of the 2009 IEEE International Conference on Intelligent Computing and Intelligent Systems, Shanghai, China, 500–503 (2009).