基于面部特征激光光点定位的人脸识别方法研究  

Face recognition based on laser light point location of facial features

作　　者：祁昌平[1] 祁昌禹[2] 万晓龙 QI Changping;QI Changyu;WAN Xiaolong(Hexi University,School of Information Technology and Media and Communications,Zhangye Gansu 734000,China;Gansu Academy of Science,Lanzhou 730000,China)

机构地区：[1]河西学院信息技术与传媒学院,甘肃张掖734000 [2]甘肃省科学院,兰州730000

出　　处：《激光杂志》2020年第10期77-81,共5页Laser Journal

基　　金：国家自然科学基金项目(No.19BYY186)。

摘　　要：人脸识别作为重要的身份鉴别技术,需要具有较好的识别时效性和准确性,为此提出一种基于面部特征激光光点定位的人脸识别方法。通过积分投影技术选择面部结构中的脸角点、嘴角点和鼻尖作为特征点。为减少面部表情变化对识别效果的影响,避开受影响最大的嘴部区域,对各特征点进行等距测地轮廓线采样,利用Procrustes分析采样点提取面部轮廓线,并选用加权平均法在二值化图像中区域化分割面部轮廓,完成激光光点定位人脸识别的全过程。实验结果表明,所提方法识别率较高、准确性更优,最高识别率为96.3%,具有较好的应用前景。As an important identification technology,face recognition needs to have better timeliness and accuracy.Therefore,a face recognition method based on laser light point location of face features is proposed.The feature points of face corner,mouth corner and nose tip are selected by integral projection technology.In order to reduce the influence of facial expression changes on the recognition effect and avoid the most affected mouth area,the contour of each feature point is sampled by isometric geodesic contour,and the facial contour is extracted by using the Proclustes analysis sampling point,and the weighted average method is used to segment the facial contour in binary image,completing the whole process of face recognition based on laser light point positioning.The experimental results show that the proposed method has higher recognition rate and better accuracy,and the highest recognition rate is 96.3%,which has a better application prospect.

关 键 词：激光光点定位 面部特征点 面部区域化分割 等距测地轮廓线 Procrustes分析 

分 类 号：TN249[电子电信—物理电子学]