Face Detection From Still Images Using Edge Information

• Main Authors: YU CHENG YI, 游正義
• Other Authors: Ouyang, Y. C.
• Format: Others
• Language: zh-TW
• Published: 2002
• Online Access: http://ndltd.ncl.edu.tw/handle/86571818766114128957

碩士 === 國立中興大學 === 電機工程學系 === 91 === Human face detection and recognition is a very popular research subject. Up to the present, it is still a hard problem to detect a human face as an image by machine, if the image contains some faces. In many practical applications (e.g. personal identification using photo-IDs), face images generally contain a simple background. In this thesis, we propose a face detection method based on the information from the edge shapes. High face detection rate can be obtain under the simple color background, posture and countenance. Input image is first enhanced by means of histogram equalization, adding some preprocess to improve the performance for our face detection system, followed by edge detection. The face contour is finally extracted using the edges information. Experimental results are showed to verify the collectness effectiveness of this method.