• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2005년도 추계종합학술대회
• /
• pp.355-358
• /
• 2005

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%.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.961-962
• /
• 2008

This paper proposes a touchless fingerprint acquisition device for five fingerprints. In conventional devices, they are focused for taking 1 touchless fingerprint image. But this device is taking 5 fingerprint image at a time. Also, it considers problems for focus and rolling in touchless fingerprint recognition for using camera. They affects fingerprint recognition, so this research is performed to measure and analyze these problems.

• 한국IT서비스학회지
• /
• 제6권2호
• /
• pp.161-175
• /
• 2007

Face recognition technology has lately attracted considerable attention because of its non-intrusiveness, usability and applicability. Related companies insist that their commercial products show the recognition rates more than 95% according to their self-testing. But, the rates cannot be admitted as official recognition rates. So, performance evaluation methods and tools are necessary to objectively measure the accuracy and performance of face recognition systems. In this paper, I propose a reference model for biometrics recognition evaluation tools, and implement an evaluation tool for the face recognition system based on the proposed reference model.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2007년도 하계종합학술대회 논문집
• /
• pp.287-288
• /
• 2007

In this paper, we propose a fake-iris detection method. In order to detect the fake-iris, we measure physiological features which are the reflectance ratio of the iris to the sclera at 750 nm and that at 850nm. In order to classify live and fake iris features, we use support vector data description (SVDD). From our experimental results, it is clear that our fake-iris detection method achieves high performance when distinguishing between a live-iris and a fake-iris.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2004년도 심포지엄 논문집 정보 및 제어부문
• /
• pp.114-116
• /
• 2004

Biometrics consist of technologies that support automatic identification or verification of identity based on behavioral or physical traits. Biometrics can authenticate identities since they measure unique individual characteristics including fingerprints, hand geometry, iris, hand vascular patterns and facial characteristics. we review the state of the hand vascular patterns identification technology and compare other competitive authentication technologies such as cryptography, electronic signature and PKI.

• 한국정보통신학회:학술대회논문집
• /
• 한국정보통신학회 2018년도 춘계학술대회
• /
• pp.472-474
• /
• 2018

최근 급속한 데이터의 생성으로 인하여 빅데이터 기술이 발전하고 있다. 특히 생체신호를 측정하는 웨어러블 디바이스의 발전으로 인하여 다양한 생체신호가 기하급수적으로 증가하고 있다. 그래서 기하급수적으로 증가한 생체신호의 특징을 파악하여 체계적으로 저장하는 저장소 기술이 필요하다. 본 논문은 생체신호의 특징을 파악하여 생체신호를 저장하는 저장소 설계와 생체신호를 수집하는 기술을 연구하고자 한다.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2008년도 하계종합학술대회
• /
• pp.863-864
• /
• 2008

To predict the iris matching performance and guarantee its reliability, image quality measure prior to matching is desired. An analysis on iris image degradation factors which deteriorate matching performance is a basic step for iris image quality measure. We considered five degradation factors-white-out, black-out, noise, blur, and occlusion by specular reflection-which happen generally during the iris image acquisition process. Experimental results show that noise and white-out degraded the EER most significantly, while others on EER were either insignificant or degradation images resulted in even better performance in some cases of blur. This means that degradation factors that affect the performance can be different from those based on human perception or image degradation evaluation.

• ETRI Journal
• /
• 제25권5호
• /
• pp.401-406
• /
• 2003

A biometric system determines the identity of a person by measuring physical features that can distinguish that person from others. Since biometric features have many variations and can be easily corrupted by noises and deformations, it is necessary to apply machine learning techniques to treat the data. When applying the conventional machine learning methods in designing a specific biometric system, however, one first runs into the difficulty of collecting sufficient data for each person to be registered to the system. In addition, there can be an almost infinite number of variations of non-registered data. Therefore, it is difficult to analyze and predict the distributional properties of real data that are essential for the system to deal with in practical applications. These difficulties require a new framework of identification and verification that is appropriate and efficient for the specific situations of biometric systems. As a preliminary solution, this paper proposes a simple but theoretically well-defined method based on a statistical test theory. Our computational experiments on real-world data show that the proposed method has potential for coping with the actual difficulties in biometrics.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2005년도 추계종합학술대회
• /
• pp.351-354
• /
• 2005

In this paper, we propose a new fake-iris detection method. In order to detect the fake-iris, we measure the change of the reflectance ratio between iris and sclera. In a live iris, the amount of change is relatively large compared to that in a fake-iris. From our experimental results, it is clear that our fake-iris detection method achieves high performance when distinguishing between a live-iris and a fake-iris.

• 대한전자공학회논문지SP
• /
• 제47권5호
• /
• pp.183-191
• /
• 2010

생체인식은 개인의 유일하면서 변화하지 않는 생체의 특징을 이용하여 개인의 본인 여부를 판별하는 방법으로써 널리 사용되어 왔다. 생체정보의 고유 불변한 특징을 저장하는 것은 개인정보의 노출에 따른 보안상의 문제점을 갖고 있으며 이를 해결하기 위해 제안된 방법이 가변생체인식 (cancelable biometrics)이다. 가변생체인식은 생체정보의 도난이나 도용으로부터 강인하며 재생성 가능한 생체템플릿을 제공하는 생체 인식방법이다. 본 논문에서는 변환 생체인식의 한 가지 방법으로써 얼굴 생체정보의 새로운 이진화 방법을 제안한다. 얼굴 생체정보의 이진화를 위한 특징추출은 얼굴정보의 부분적 변화에 강인한 영역기반 주성분 분석(Subpattern-based PCA)을 이용하였으며 이로부터 얻어진 특징을 보조정보에 기반한 방법으로 이진화 하였다. 획득된 이진비트열은 영역기반 주성분 분석의 사용으로 여러 얼굴 영역의 고려와 함께, 선택된 주성분 개수만큼의 계수들에 대한 이진화 값들을 포함하고 있다. 이러한 서로 다른 얼굴영역의 여러 주성분들에서 추출된 이진비트열중 구분력이 좋은 비트 값들을 선택하였으며, 선택된 비트 값들은 이진화를 위한 보조 정보가 노출된 경우에서도 원 얼굴특징벡터보다 향상된 인식성능을 보여준다.