Abstract
The dorsal hand vein (DHV) biometrics is commonly employed in personal verification or identification due to its excellent anti-counterfeit and liveness detection capabilities. In the field of DHV biometrics, the performance of deep convolutional neural networks (DCNNs) is limited due to insufficient labeled data, less discriminative features, and various image qualities of different dataset. In this paper, we take the vanilla ResNet50 as backbone, combine the Squeeze-and-Excitation block and adopt knowledge transfer to enhance the recognition performance. First, SE-ResNet50 model is constructed by embedding the Squeeze-and-Excitation module into each convolutional block of the vanilla ResNet50. Second, the transfer learning strategy is adopted to speed up training efficiency and enhance generalization capability. That is, the MPD palmprint dataset is employed to pre-train the SE-ResNet model. The parameters of the pre-trained model are adjusted using DHV datasets to achieve knowledge transfer. Three datasets were conducted in our experiments. The experimental results demonstrated that the adoption of attention mechanism and transfer learning can significantly improve the recognition accuracy. The proposed SE-ResNet50 model achieves competitive performance of the state of the art with higher computational efficiency and generalization capability.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
5. References
Akram MU, Awan HM, Khan AA (2014) Dorsal hand veins based person identification. In: 4th international conference on image processing theory, tools and applications (IPTA). IEEE, Istanbul, Turkey, pp 1–6
Ding YH, Zhuang DY, Wang KJ (2005) A study of hand vein recognition method. In: International conference mechatronics and automation. IEEE, Niagara Falls, Canada, pp 2106–2110
Gu GJ, Bai PR, Li H, Liu QY, Han C, Min XL, Ren YD (2021) Dorsal hand vein recognition based on transfer learning with fusion of LBP feature. In: 15th Chinese conference on biometric recognition. Springer, Shanghai, China, pp 221–230
Guo ZY, Ma Y, Min XL, Li H, Liu QY, Han C, Yang G, Bai PR, Ren YD (2020) A novel algorithm of dorsal hand vein image segmentation by integrating matched filter and local binary fitting level set model. In: 7th international conference on information science and control engineering (ICISCE). IEEE, Changsha, China, pp 81–85
He KM, Zhang XY, Ren SQ, Sun J (2016) Deep residual learning for image recognition. In: Proceedings of the IEEE Conference on Computer Vision and pattern recognition. IEEE, pp 770–778
Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. In: Proceedings of the IEEE Conference on Computer vision and pattern recognition. IEEE, pp 7132–7141
Huang D, Zhang RK, Yin Y, Wang YD, Wang YH (2017) Local feature approach to dorsal hand vein recognition by centroid-based circular key-point grid and fine-grained matching. Image vis Comput 58:266–277
Jia W, Xia W, Zhang B, Zhao Y, Fei LK, Kang WX, Huang D, Guo GD (2021) A survey on dorsal hand vein biometrics. Pattern Recogn 120:108–122
Jia D, Wei D, Richard S, Li JL, Kai L, Li FF (2009) Imagenet: a large-scale hierarchical image database. In: Proceedings of the IEEE Conference on Computer Vision and pattern recognition. IEEE, pp 248–255
Joshua N, Stephen E, Bhattacharyya D, Chakkravarthy M, Kim HJ (2021) Lung cancer classification using squeeze and excitation convolutional neural networks with grad cam++ class activation function. Traitement du Signal 38(4)
Kauba C, Uhl A (2018) Shedding light on the veins-reflected light or transillumination in hand-vein recognition. In: 11th IAPR international conference on biometrics (ICB). IEEE, Queensland, Australia, pp 283–290
Kuzu RS, Maiorana E, Campisi P (2020) Vein-based biometric verification using transfer learning. In: 43rd international conference on telecommunications and signal processing (TSP). IEEE, Milan, Italy, pp 403–409
Lee JC, Lo TM, Chang CP (2016) Dorsal hand vein recognition based on directional filter bank. SIViP 10(1):145–152
Li XX, Huang D, Wang YH (2016) Comparative study of deep learning methods on dorsal hand vein recognition. In: 10th Chinese conference on biometric recognition. Springer, Tianjin, China, pp 296–306
Liu F, Jiang SK, Kang B, Hou T (2020) A recognition system for partially occluded dorsal hand vein using improved biometric graph matching. IEEE Access 74525–74534
Liu Y, Zhang W, Wang J, Wang JY (2021) Data-free knowledge transfer: a survey. arXiv:2112.15278
Mnih V, Heess N, Graves A (2014) Recurrent models of visual attention. In: Advances in Neural Information Processing Systems, vol 27
Rajalakshmi M, Annapurani K (2022) A deep learning based palmar vein recognition: transfer learning and feature learning approaches. In: Proceedings of international conference on deep learning, computing and intelligence. Springer, Singapore, pp 581–591
Ribaric S, Ariyaeeinia A, Pavesic N (2016) De-identification for privacy protection in multimedia content: a survey. Signal Process Image Commun 47:131–151
Roy AG, Navab N, Wachinger C (2018) Concurrent spatial and channel ‘squeeze & excitation’ in fully convolutional networks. In: International conference on medical image computing and computer-assisted intervention. Springer, Cham, pp 421–429
Simonyan K, Zisserman A (2014) Very deep convolutional networks for large-scale image recognition. arXiv:1409.1556
Uhl A, Busch C, Marcel S, Veldhuis RNJ (2020) Handbook of vascular biometrics. Springer Nature
Vairavel KS, Ikram N, Mekala S (2019) Performance analysis on feature extraction using dorsal hand vein image. Soft Comput 23(18):8349–8358
Wang J, Wang GQ, Zhou M (2017) Bimodal vein data mining via cross-selected-domain knowledge transfer. IEEE Trans Inf Forensics Secur 13(3):33–744
Xie S, Girshick R, Dollár P, Tu Z, He K (2017) Aggregated residual transformations for deep neural networks. In: Proceedings of the IEEE conference on computer vision and pattern recognition. IEEE, pp 1492–1500
Yosinski J, Clune J, Bengio Y, Lipson H (2014) How transferable are features in deep neural networks? In: Advances in Neural Information Processing Systems, vol 27
You KC, Liu Y, Wang JM, Long MS (2021) Logme: practical assessment of pre-trained models for transfer learning. In: International conference on machine learning. PMLR, pp 12133–12143
Zhang YY, Zhang L, Zhang RX, Li SX, Li JL, Huang FY (2020) Towards palmprint verification on smartphones. arXiv:2003.13266
Zhang D, Guo ZH, Gong YZ (2016) Dorsal hand recognition. In: Multispectral biometrics. Springer, Cham, pp 165–186
Zhou BL, Khosla A, Lapedriza A, Oliva A, Torralba A (2016) Learning deep features for discriminative localization. In: Proceedings of the IEEE Conference on Computer Vision and pattern recognition. IEEE, pp 2921–2929
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Zong, C. et al. (2023). An Efficient Model for Dorsal Hand Vein Recognition Based on Combination of Squeeze-and-Excitation Block and Vanilla ResNet. In: Yadav, S., Kumar, H., Kankar, P.K., Dai, W., Huang, F. (eds) Proceedings of 2nd International Conference on Artificial Intelligence, Robotics, and Communication . ICAIRC 2022. Lecture Notes in Electrical Engineering, vol 1063. Springer, Singapore. https://doi.org/10.1007/978-981-99-4554-2_21
Download citation
DOI: https://doi.org/10.1007/978-981-99-4554-2_21
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-4553-5
Online ISBN: 978-981-99-4554-2
eBook Packages: Intelligent Technologies and RoboticsIntelligent Technologies and Robotics (R0)