A Study on Face Detection Using Template Matching and Elliptical Information

템플릿과 타원정보를 이용한 얼굴검출에 관한 연구

  • 강우석 (연세대학교 전기공학과) ;
  • 김현술 (연세대학교 전기공학과) ;
  • 박남준 (연세대학교 전기공학과) ;
  • 박상희 (연세대학교 전기공학과)
  • Published : 1998.11.28

Abstract

This paper proposes a new segmentation method of human races from grey scale images with clutter using a racial template and elliptical structure of the human head. Face detection technique can be applied in many areas of image processing such as face recognition, composition and computer graphics. Until now, many researches about face detection have been conducted, and applications in more complicated conditions are increasing. The general case is more in a complicated background than in a simple one, and a image with not only one face. Research and development of face detection in such a general case are growing rapidly, and the necessity for that is increasing continuously. Sirohey proposed a face detection method using linearized elliptical equation. The method designed in this paper is improved to be applicable even in the more general cases like where the face is much smaller than the image size and with many faces in one image using template matching and elliptic fitting technique.

Keywords