A Study on Deep Learning-based Method for Finger Scanning Authentication
0
조회
0
다운로드
0
인용
0
참고문헌
키워드
초록
Finger vein is considered to be an extremely secure and reliable biometric modality used for Authentication. The finger vein authentication is the recently developing biometric technology which gains more attention from the past decade. The traditional methods of finger vein authentication techniques are affected by several factors such as illumination variations and image misalignments lead to inefficient feature extraction. Therefore, the overall performance of the authentication system is reduced. A Convolutional Neural Network is a deep transfer learning-based approach which overcomes the issues in the traditional finger vein authentication. A deep learning based feature extraction technique for digit vein images is proposed in this paper. The Convolution Neural Network (CNN) proposed in this paper contains six groups of a convolutional layer, RELU, batch normalization, and max pooling layer, followed by a abundantly coupled layer and a flatten layer. The CNN is tested, trained using a different group of images from the SDUMLA finger vein database and overall accuracy of 99.84%, 0.00081 of FAR and 0.04 of FRR were achieved.
Abstract
Finger vein is considered to be an extremely secure and reliable biometric modality used for Authentication. The finger vein authentication is the recently developing biometric technology which gains more attention from the past decade. The traditional methods of finger vein authentication techniques are affected by several factors such as illumination variations and image misalignments lead to inefficient feature extraction. Therefore, the overall performance of the authentication system is reduced. A Convolutional Neural Network is a deep transfer learning-based approach which overcomes the issues in the traditional finger vein authentication. A deep learning based feature extraction technique for digit vein images is proposed in this paper. The Convolution Neural Network (CNN) proposed in this paper contains six groups of a convolutional layer, RELU, batch normalization, and max pooling layer, followed by a abundantly coupled layer and a flatten layer. The CNN is tested, trained using a different group of images from the SDUMLA finger vein database and overall accuracy of 99.84%, 0.00081 of FAR and 0.04 of FRR were achieved.
참고문헌 (20)
- Renshan Jia, "A Student Based Biometric Application", 아시아태평양융합연구교류논문지, 5(4), pp.103~112 (2019), 사단법인 한국융합기술연구학회[DOI]
- N. Miura, "Feature mining of finger vein patterns based on iterative line tracking and its application to personal identification", Systems and Computers, 35(7), pp.194~203 (2004)
- N. Miura, "Mining of Finger-Vein Patterns Using Maximum Curvature Points in Image Profiles", IEICE Trans. Inf. Syst., E90-D(8), pp.1185~1194 (2005)
- F. Cui, "Level Fusion of Fingerprint and Finger Vein Recognition", Journal of Computational Information Systems, 7(16), pp.5723~5731 (2011)
- V. Thien, "Palmprint verification using GridPCA for Gabor features", Proceedings of the 2011 Symposium on Information and Communication Technology (2011)
- Lei Li, "An Extensive Review on Recent Deep Learning Applications", 아시아태평양융합연구교류논문지, 5(3), pp.221~231 (2019), 사단법인 한국융합기술연구학회[DOI]
- 노먼 칸, "Efficient Sport Videos Classification via Convolutional Neural Network", 한국차세대컴퓨팅학회 논문지, 16(6), pp.7~16 (2020), 한국차세대컴퓨팅학회[DOI]
- 이민경, "컨볼루션 신경망의 신뢰도 규제를 이용한 CT 영상에서 악성 난소 종양 판별 방법", 한국차세대컴퓨팅학회 논문지, 16(2), pp.80~87 (2020), 한국차세대컴퓨팅학회[DOI]
- W. Liu, "Finger vein recognition based on deep learning", 12th IEEE Conference on Industrial Electronics and Applications (ICIEA) (2017)
- H. G. Hong, "Convolutional neural network-based fingervein recognition using NIR image sensors", Sensors, 17(6), pp.1297~1318 (2017)
- W. C. Yen, "Data Coherence Problem in a Multicache System", IEEE Trans. on Soft Eng, SE‐34(1), pp.56~65 (1985)
- Deepika Dubey, "Echelon Based Pose Generalization of Facial Images Approaches", 아시아태평양융합연구교류논문지, 3(1), pp.63~75 (2017), 사단법인 한국융합기술연구학회[DOI]
- Y. Zhang, "Adaptive Gabor Convolutional Neural Networks for Finger-Vein Recognition", International Conference on High Performance Big Data and Intelligent Systems(HPBD&IS), pp.219~222 (2019)
- Boucherit, M. O., "Finger vein identification using deeply-fused Convolutional Neural Network", Journal of King Saud University – Computer and Information Sciences (2020)
- "The SDUMLA-HMT database"[URL]
- 장준수, "합성곱 신경망을 이용한 요추 특징점 검출에 관한 연구", 차세대컨버전스정보서비스기술논문지, 9(3), pp.263~272 (2020), 차세대컨버전스정보서비스학회[DOI]
- 박예랑, "MR 영상을 이용한 합성곱 신경망 기반의 상완골두 및 관절와 자동 분할", 차세대컨버전스정보서비스기술논문지, 9(1), pp.95~102 (2020), 차세대컨버전스정보서비스학회[DOI]
- 배한준, "순환신경망을 이용한 자기장 기반 실내측위시스템", 한국차세대컴퓨팅학회 논문지, 14(6), pp.57~65 (2018), 한국차세대컴퓨팅학회
- 최현종, "이중흐름 3차원 합성곱 신경망 구조를 이용한 효율적인 손 제스처 인식 방법", 한국차세대컴퓨팅학회 논문지, 14(6), pp.66~74 (2018), 한국차세대컴퓨팅학회
- 김민준, "딥 러닝 기반 설명 가능한 인공지능과 무인기의 다중 이미지 센서를 활용한 무 시들음병 탐지 프레임워크", 한국차세대컴퓨팅학회 논문지, 17(1), pp.7~15 (2021), 한국차세대컴퓨팅학회[DOI]