Proceedings of the IEEK Conference (대한전자공학회:학술대회논문집)

The Institute of Electronics and Information Engineers (대한전자공학회)
권호 표지

Liveness Detection of Fingerprints Using Correlation Filters

상관 필터를 이용한 위조 지문 검출 방법

  • Choi, Hee-Seung (Department of Electrical and Electronic Engineering, Yonsei University, Biometrics Engineering Research Center) ;
  • Choi, Kyung-Taek (Department of Electrical and Electronic Engineering, Yonsei University, Biometrics Engineering Research Center) ;
  • Kim, Jai-Hie (Department of Electrical and Electronic Engineering, Yonsei University, Biometrics Engineering Research Center)
  • 최희승 (연세대학교 공과대학 전기전자공학과 생체인식 연구센터) ;
  • 최경택 (연세대학교 공과대학 전기전자공학과 생체인식 연구센터) ;
  • 김재희 (연세대학교 공과대학 전기전자공학과 생체인식 연구센터)
  • Published : 2005.11.26
Download PDF
⟨ Previous Next ⟩

Abstract

Fingerprint recognition systems are the most widely used in biometrics for personal authentication. As they become more familiar, the security weaknesses of fingerprint sensors are becoming better known. In this paper, we propose a liveness detection method that applies correlation filter to the fingerprint recognition systems. The physiological characteristic of sweat pore, observed only in live people, is used as a measure to classify 'live' fingers from 'spoof' fingers. Previous works show that detection of sweat pores and perspiration patterns in fingerprint images can be used as an anti-spoofing measure. These methods don't consider the characteristic of pores in each individual. We construct the correlation filters of each individual which are composed of their pore information. We make the final decision about the "livens" of fingerprint using correlation output. The proposed algorithm was applied to a data set of 110 live, 110 spoof fingerprint images from optical fingerprint scanner and achieved classification rate of 80%.

Keywords