人眼定位算法与人脸图像标准化

提出了一种改进的人脸图像标准化算法.该算法先利用改进的人脸图像的垂直灰度积分投影确定脸的左右边界,再根据人脸图像的水平灰度投影曲线来确定眼睛的大致高度.利用由上述方法得到的参数对人脸进行第一次裁剪,然后对人眼精确定位,最后对图像进行旋转、剪切和缩放,获得人脸的标准图像.采用ORL人脸数据库进行仿真实验,结果表明,该算法消除了背景、头发等干扰,保留了人脸识别所需的主要信息,有利于后续的特征提取与识别.

Method of human eyes locating and face image standardization

An improved method of face image standardization is presented.The right and left borderline of face is approximately determined according to improved vertical graylevel projection,and the height of human eyes is also approximately determined according to horizontal graylevel projection.The parameters obtained above are used for the first cropping of face image.Then the accurate positions of human eyes are located.Finally,by cropping,rotating and zooming the image,a standard image can be achieved. Experimental results based on ORL face database demonstrate that interferences such as backgrounds,hair and costumes are almost all eliminated,while the main information required in face recognition is preserved.The method is much beneficial for the followup feature extraction and recognition.