한국정보처리학회:학술대회논문집 (Proceedings of the Korea Information Processing Society Conference)

한국정보처리학회 (Korea Information Processing Society)
권호 표지

임베디드 시스템을 위한 눈 찾기 알고리즘

A Simple Eye Detection Algorithm for Embedded System

  • 이영재 (안동대학교 컴퓨터 공학과) ;
  • 김익동 (안동대학교 컴퓨터 공학과) ;
  • 최미순 (안동대학교 컴퓨터 공학과) ;
  • 심재창 (안동대학교 컴퓨터 공학과)
  • Lee Yung-Jae (Department of Computer Engineering, Andong National University) ;
  • Kim Ik-Dong (Department of Computer Engineering, Andong National University) ;
  • Choi Mi-Soon (Department of Computer Engineering, Andong National University) ;
  • Shim Jae-Chang (Department of Computer Engineering, Andong National University)
  • 발행 : 2004.11.01
PDF 다운로드
⟨ 이전 논문 다음 논문 ⟩

초록

Many of facial feature extracting applications and systems have been developed in the field of face recognition systems and its application, and most of them use the eyes as a key-feature of human face. In this paper we show a simple and fast eye detection algorithm for embedded systems. The eyes are very important facial features because of the attribution they have. For example, we know the darkest regions in a face are the pair of pupils, and the eyes are always a pair and parallel. Using such attributors, our algorithm works well under various light conditions, size of face in image, and various pose such as panning and tilting. The main keys to develop this algorithm are the eyes' attribution that we can usually contemplate and easily find when we think about what is the attribution that the eyes have. With some constraints of the eyes and knowledge of the anthropometric human face, we detect human eye in an image, and the experimental results demonstrate successful eye detection.

키워드