GAs를 이용한 Hand-Geometry의 특징 추출 알고리즘 및 영상 획득 시스템 구현에 관한 연구

Abstract

Biometrics is getting more and more attention in recent years for security and other purpose. So far, only fingerprint has seen limited success for on-line security check, since other biometrics verification and identification systems require more complicated and expensive acquisition interfaces and recognition processes.

Hand-Geometry has been used for biometric verification and identification because of its acquisition convenience and good verification and identification performance. Therefore, this paper propose Hand-Geometry recognition system based on geometrical of hand. From anatomical point of view, human hand can be characterized by its length, width, thickness, geometrical composition, shapes of the palm, and shape and geometry of the fingers. Unlike palmprint verification Hand-Geometry does not involve extraction of detailed features of the hand(for example, wrinkles on the skin).

Whole system is consisted of image acquisition part, processing part, actuator part. Image acquisition part is consisted of image capture board and CCD camera that is image acquisition system. Processing part extracts feature points in hand image that inputted from CCD camera using GAs that imitates nature evolution and has excellent performance in search algorithm. And after extract feature points, image of inputted color scale changes to gray scale, and extracts characteristic data. Finally, feature data that is gotten from processing part is transmitted by printer port and confirmed result of Hand-Geometry recognition through actuator part. This paper proposes Hand-Geometry recognition system having with function such as upside. This system presents verification results based on hand measurements of 100 data about 20 individuals captured over real time. The recognition process has been tested on a size of 320 × 240 image, and result of the recognition process have hit rate of 94％ and FAR of 0.021.