Fingerprint Spoof Detection Using Contrast Enhancement and Convolutional Neural Networks
Recently, as biometric technology grows rapidly, the importance of fingerprint spoof detection technique is emerging. In this paper, we propose a technique to detect forged fingerprints using contrast enhancement and Convolutional Neural Networks (CNNs).
- PDF / 697,971 Bytes
- 8 Pages / 439.37 x 666.142 pts Page_size
- 45 Downloads / 229 Views
Abstract. Recently, as biometric technology grows rapidly, the importance of fingerprint spoof detection technique is emerging. In this paper, we propose a technique to detect forged fingerprints using contrast enhancement and Convolutional Neural Networks (CNNs). The proposed method detects the fingerprint spoof by performing contrast enhancement to improve the recognition rate of the fingerprint image, judging whether the sub-block of fingerprint image is falsified through CNNs composed of 6 weight layers and totalizing the result. Our fingerprint spoof detector has a high accuracy of 99.8% on average and has high accuracy even after experimenting with one detector in all datasets. Keywords: Biometrics · Fingerprint spoof detection neural networks · Multimedia security
1
·
Convolutional
Introduction
Fingerprint recognition is an automation technique that proves whether two human fingerprints match. Fingerprint recognition scans human fingerprints that have different shapes for each person in a short period of time, and then releases security or other functions if it is determined by the fingerprint of the same user [1]. However, fingerprint recognition technology has a problem of leakage of biometric information. Biometric information including fingerprints are unique information that can not be changed. If leaked once, malicious users may impersonate and threaten security [2]. The problem of leakage of biometric information is a serious security threat, and the importance of technology for verifying actual biometric information is rapidly increasing. Hence, in the fingerprint recognition system, it is essential to distinguish whether the fingerprint to be authenticated is an alive part of a person or a forged fingerprint. The fingerprint spoof detection technique is divided into hardware and software techniques depending on whether additional sensors are used or not [3]. Among them, the software technique has a merit that it can be used in a general c Springer Nature Singapore Pte Ltd. 2017 K. Kim and N. Joukov (eds.), Information Science and Applications 2017, Lecture Notes in Electrical Engineering 424, DOI 10.1007/978-981-10-4154-9 39
332
H.-U. Jang et al.
fingerprint recognition device because it judges whether or not the fingerprint is falsified by using the fingerprint image. The software techniques can be classified as feature based and deep learning based. The feature based detection method is mainly a research to discriminate fingerprint liveness with a single feature point in early stage [1,3], and it has not been shown good performance for various fake materials [4,5]. Feature based detection with multiple features was proposed to detect various fingerprints liveness, and Dubey et al. used the Speeded-Up Robust Features (SURF), Pyramid Histogram of Oriented Gradients (PHOG), and texture features to detect fingerprint liveness [4]. Rattani et al. proposed an automatic adaptation of a liveness detector to new spoof materials [6]. They used Gray Level Co-occurence Matrix (GLCM), Histogram of Oriented Gradients (HOG),
Data Loading...
