期刊名称：International Journal of Computer Science and Network Security
印刷版ISSN：1738-7906
出版年度：2006
卷号：6
期号：5A
页码：132-135
出版社：International Journal of Computer Science and Network Security
摘要：This paper proposed a neural network-based face segmentation method for eyes location. After face is detected, it is segmented automatically by a competitive Hopfield neural network (CHNN) and the facial feature candidates such as eyes, eyebrows and mouth are obtained, and then eyes are located by facial features evaluation and validation, which is based on face’s geometrical structures. To enhance the difference between facial features and other face area, the color face image is converted into a color-ratio grey image before face segmentation. Experimental results show that the proposed system performs well.
关键词：Competitive Hopfield neural network, Eyes location, Face segmentation, color-ratio image
