一种基于多分量信息融合的人脸检测方法  

An Approach of Face Detection Based on Multicomponent Information Fusion

作　　者：陈健[1] 周利莉[1] 史红刚[1] 

机构地区：[1]信息工程大学信息科学系,郑州450002

出　　处：《计算机工程与应用》2005年第36期57-59,65,共4页Computer Engineering and Applications

摘　　要：人脸检测是全自动人脸识别系统和许多监视系统的基础,在许多领域有着广泛的应用。文章提出了一种基于多分量信息融合的人脸检测方法。首先进行彩色空间的变换,检测出图像中的肤色区域;建立眼睛颜色模型,并根据眼睛﹑嘴在不同分量上的分布特征,将它们提取出来;最后融合眼睛﹑嘴候选区域的信息,利用特征不变的方法进行人脸的确定。实验结果表明,该方法能够快速有效地检测出人脸,并能够确定眼﹑嘴的位置。Face detection is the first step for any fully automatic face recognition system and in many surveillance systems,it has wide applications in many fields.This paper presents an approach of face detection based on multicomponent information fusion,First,color space transformation is used to detect skin color regions,then we construct an eye color model and extract eye and mouth candidates based on the distribution features on each component of color space.Finally we fuse the information of each candidate and validate faces using feature invariant method, Experimental results show that this algorithm is effective in detecting faces as well as localizing mouth and eyes.

关 键 词：人脸检测 颜色模型 信息融合 特征不变 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]