A Face Detection Method Based on Color and Geometry Information

摘要：

A face detection method based on color and geometry information is researched in this paper. Firstly, the color image with a few faces is pre-proceeded by light compensation. Secondly, the color image proceeded is segmented based on the skin color information to get a binary human face image. At last, the skin areas are marked, then, the ratio of width and height, the number of width pixels and height pixels are used to select right human face areas. The experimental results show that this method considers color and geometry information comprehensively, the effect of face detection is fine, the method is easy and the computation speed is fast.

关键词： Face detection Color space Noise smoothing light compensation

作者: Rong Wang XiaoGang Yang

作者单位: College of security & protection, Chinese peoples public security university, Beijing 100038

会议类型: 国际会议

会议名称: The 24th Chinese Control and Decision Conference (第24届中国控制与决策学术年会 2012 CCDC)

会议地点: 太原

会议语种:英文

页码: 1972-1975

在线出版日期: 2012-05-23（万方平台首次上网日期，不代表论文的发表时间）