Journal of Sensor Science and Technology (센서학회지)
- Volume 8 Issue 1
- /
- Pages.63-68
- /
- 1999
- /
- 1225-5475(pISSN)
- /
- 2093-7563(eISSN)
Eye Pattern Detection Using SVD and HMM Technique from CCD Camera Face Image
CCD 카메라 얼굴 영상에서의 SVD 및 HMM 기법에 의한 눈 패턴 검출
- Jin, Kyung-Chan (Sensor Technology Research Center, Kyungpook National Univ.) ;
- Miche, Pierre (INSA De Rouen, France) ;
- Park, Il-Yong (School of Electronic & Electrical Eng., Kyungpook National Univ.) ;
- Sohn, Byung-Gi (School of Electronic & Electrical Eng., Kyungpook National Univ.) ;
-
Cho, Jin-Ho
(School of Electronic & Electrical Eng., Kyungpook National Univ.)
- Published : 1999.01.31
Abstract
We proposed a method of eye pattern detection in the 2-D image which was obtained by CCD video camera. To detect face region and eye pattern, we proposed pattern search network and batch SVD algorithm which had the statistical equivalence of PCA. We also used HMM to improve the accuracy of detection. As a result, we acknowledged that the proposed algorithm was superior to PCA pattern detection algorithm in computational cost and accuracy of defection. Furthermore, we evaluated that the proposed algorithm was possible in real-time face pattern detection with 2 frame images per second.
CCD 카메라에 얻어진 비디오 신호로부터 디지털화된 얼굴영상을 얻은 다음 신속하게 눈 패턴을 검출할 수 있는 알고리즘을 제안하였다. 얼굴영역 및 눈 패턴 검출을 위해 주축성분분석(Principle Component Analysis : PCA)과 통계적인 성질이 유사한 일괄처리 SVD(Singular Value Decomposition)알고리즘 및 패턴서치회로망을 이용하였고, 정확도를 향상하기 위해 HMM(Hidden Markov Model)을 이용하였다. 제안한 알고리즘으로 기존의 주축성분분석을 이용한 패턴 검출보다 계산량이 적고 정확도도 우수하며 초당 2 프레임 정도의 실시간 처리가 가능함을 알 수 있었다.
Keywords