Presentation 2022-10-03
Performance Improvement of CNN-Based Fingerprint Recognition Using Multiple Attention Mechanism
Nagisa Sasuga, Koichi Ito, Takafumi Aoki,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Fingerprint recognition methods that extract multiple features from a fingerprint image using a Convolutional Neural Network (CNN) have been proposed. While this method can achieve high accuracy with a small number of training data, there is the problem that the network structure becomes complex and the size of the network becomes large. In this paper, we investigate a method to improve the accuracy of fingerprint recognition without increasing the network size by introducing multiple attentional mechanisms into the CNN. We demonstrate the effectiveness of the proposed method through performance evaluation experiments using public datasets.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) biometrics / fingerprint recognition / convolutional neural network / attention mechanism
Paper # BioX2022-55
Date of Issue 2022-09-26 (BioX)

Conference Information
Committee BioX
Conference Date 2022/10/3(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair Hitoshi Imaoka(NEC)
Vice Chair Norihiko Okui(KDDI Research) / Naoyuki Takada(SECOM)
Secretary Norihiko Okui(NEC) / Naoyuki Takada(MitsubishiElectric)
Assistant Hiroyuki Suzuki(Gunma Univ) / Shinichi Shirakawa(Yokohama National Univ.)

Paper Information
Registration To Technical Committee on Biometrics
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Performance Improvement of CNN-Based Fingerprint Recognition Using Multiple Attention Mechanism
Sub Title (in English)
Keyword(1) biometrics
Keyword(2) fingerprint recognition
Keyword(3) convolutional neural network
Keyword(4) attention mechanism
1st Author's Name Nagisa Sasuga
1st Author's Affiliation Tohoku University(Tohoku Univ.)
2nd Author's Name Koichi Ito
2nd Author's Affiliation Tohoku University(Tohoku Univ.)
3rd Author's Name Takafumi Aoki
3rd Author's Affiliation Tohoku University(Tohoku Univ.)
Date 2022-10-03
Paper # BioX2022-55
Volume (vol) vol.122
Number (no) BioX-197
Page pp.pp.1-6(BioX),
#Pages 6
Date of Issue 2022-09-26 (BioX)