Title: Dual-channel heterogeneous network for finger vein recognition based on transfer learning and improved coordinate attention

Authors: Yuchuan Chen; Jing Jie; Jiahui Chai; Hui Zheng; Xiaoli Wu

Addresses: School of Automation and Electrical Engineering, Zhejiang University of Science and Technology, Liuhe Road 318#, Hangzhou, Zhejiang, 310023, China ' School of Automation and Electrical Engineering, Zhejiang University of Science and Technology, Liuhe Road 318#, Hangzhou, Zhejiang, 310023, China ' School of Automation and Electrical Engineering, Zhejiang University of Science and Technology, Liuhe Road 318#, Hangzhou, Zhejiang, 310023, China ' School of Automation and Electrical Engineering, Zhejiang University of Science and Technology, Liuhe Road 318#, Hangzhou, Zhejiang, 310023, China ' School of Automation and Electrical Engineering, Zhejiang University of Science and Technology, Liuhe Road 318#, Hangzhou, Zhejiang, 310023, China

Abstract: In order to effectively improve the recognition accuracy of finger veins, a dual-channel heterogeneous network based on transfer learning and coordinate attention is proposed. Firstly, a VGG16 and a DenseNet121 pretrained by ImageNet are selected to form the dual-channel heterogeneous network, and then the new network is trained again by finger vein image set to validly improve its characterisation power without too much training cost. Secondly, an improved coordinate attention is introduced to help the network evaluate the importance of information in different spatial locations, which contributes to outputting valid finger vein features to the classification layer. Finally, the network performance was tested based on a 700-type of finger vein image set. The experimental results show that transfer learning and improved coordinate attention validly enhance the recognition effect with low computation complexity, and the proposed network shows excellent performance for finger vein recognition.

Keywords: finger vein recognition; transfer learning; coordinate attention; convolutional neural network; feature fusion.

DOI: 10.1504/IJCSM.2024.139087

International Journal of Computing Science and Mathematics, 2024 Vol.19 No.4, pp.299 - 317

Received: 19 Sep 2023
Accepted: 27 Mar 2024
Published online: 12 Jun 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article