| Summary: | 碩士 === 中華大學 === 資訊工程學系碩士班 === 91 === In this thesis, we develop a real time system for detection of face and eyes in the video sequences. The result that may be applied to human machine interaction, surveillance (face authentication), etc. We have adopted HSI color model to extract skin-color pixels and group them by the region growing
algorithm and selected the maximum region as the face. Furthermore, we have proposed a Face-Circle Fitting (FCF) method to confirm the exacted face region. Once the face is detected, we used Dark-Pixel Filter (DPF) to perform the eye detection. When eyes are detected, we can obtain the face
location, the face tilt angle, face size and to predict the distance between a human and the camera. These information are vary useful for further applications. The proposed system has been implemented on personal computer with a PC-camera and can perform face detection and eyes detection in real-time.
key words: face detection, eye detection.
|
|---|
