@inproceedings{85bf2da281984257b926521731651273,
title = "Different Dimension Issues in Deep Feature Space for Finger-Vein Recognition",
abstract = "Hand-crafted approaches were the dominating solutions and recently, more convolutional neural network (CNN)-based methods have been proposed for finger-vein recognition. However, the previous deep learning methods usually designed the network architecture with increasing layers and parameters, which incurs device memory issues and processing speed issues. Although many researchers have devoted to design image enhancement algorithms to improve the recognition performance of hand-crafted methods, it is interesting to investigate whether deep learning method can achieve satisfactory performance without image enhancement. This paper focuses on two different dimension issues: lightweight CNN design and the impact of image enhancement on deep learning methods. The experimental results demonstrate that the proposed method LFVRN is comparable or superior to the prior competition winners. In addition, image enhancement is validated not inevitable for the proposed lightweight CNN model LFVRN.",
keywords = "CNN-based method, Finger-vein recognition, Image enhancement, Network architecture design",
author = "Yiqi Zhong and Jiahui Li and Tingting Chai and Shitala Prasad and Zhaoxin Zhang",
note = "Publisher Copyright: {\textcopyright} 2021, Springer Nature Switzerland AG.; 15th Chinese Conference on Biometric Recognition, CCBR 2021 ; Conference date: 10-09-2021 Through 12-09-2021",
year = "2021",
doi = "10.1007/978-3-030-86608-2\_33",
language = "英语",
isbn = "9783030866075",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Science and Business Media Deutschland GmbH",
pages = "295--303",
editor = "Jianjiang Feng and Junping Zhang and Manhua Liu and Yuchun Fang",
booktitle = "Biometric Recognition - 15th Chinese Conference, CCBR 2021, Proceedings",
address = "德国",
}