单人图象中人眼的快速定位  

Fast Location of Eye Features in a Single Person Image

作　　者：衣晓飞[1] 陈福接[1] 

机构地区：[1]国防科技大学计算机学院,湖南长沙410073

出　　处：《计算机工程与科学》2001年第3期15-17,24,共4页Computer Engineering & Science

摘　　要：人眼的识别是计算机人脸识别和智能监控中的重要部分。本文提出了一种基于灰度投影曲线和边缘检测进行人眼定位的新算法。首先利用原图的垂直灰度投影曲线确定人脸的左右边界 ,之后利用人脸区域的水平灰度投影确定头顶至鼻中部的上下边界 ,然后利用人脸比例模型估计人眼的大概位置 ,分析人眼区域的边缘图象 ,给出人眼的确切位置。实现表明 ,该算法简单、快速 。Eye recognition is very important in face recognition and intelligent surveillance.In this paper a new method of locating eye features in facial images based on gray level projection is put forward.First,the left and right boundaries of face is localized with vertical gray level projection.Secondly,face region's horizontal gray level projection is used to localize the top of the head and the middle of the nose.Finally,the face structure model is utilized to estimate the positions of eyes,and the edge features are checked to give the proper positions of eyes.Experiments show that the algorithm is simple,fast and robust.[WT5HZ]

关 键 词：边缘检测 人脸识别 人眼识别 计算机 彩色图象 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]