首页 | 本学科首页   官方微博 | 高级检索  
     

基于OpenCV人眼定位的人脸检测方法
引用本文:汪明明,林小竹. 基于OpenCV人眼定位的人脸检测方法[J]. 北京石油化工学院学报, 2012, 20(4): 48-53
作者姓名:汪明明  林小竹
作者单位:北京化工大学信息科学与技术学院,北京100029;北京石油化工学院信息工程学院,北京102617;北京石油化工学院信息工程学院,北京,102617
摘    要:如何准确地定位人脸是人脸识别中非常关键而且重要的一步。笔者提出一种基于OpenCV人眼定位的人脸检测方法。利用从摄像头得到的视频图像中随机获取的单帧图像作为待处理人脸图像,通过人眼定位、图像旋转、分割以及缩放得到标准化后的人脸图像,再经过ART2神经网络的学习认知,计算其识别率。对比实验说明此方法能够应用于人脸识别,并具有较快的识别速度和良好的识别率,特别方便于实际应用。

关 键 词:人眼定位  人脸检测  视频图像  神经网络  人脸识别

The Method of Face Detection Based on Eye Location Using OpenCV
Wang Mingming,Lin Xiaozhu. The Method of Face Detection Based on Eye Location Using OpenCV[J]. Journal of Beijing Institute of Petro-Chemical Technology, 2012, 20(4): 48-53
Authors:Wang Mingming  Lin Xiaozhu
Affiliation:1.College of Information Science and Technology,Beijing University of Chemical Technology,Beijing 100029,China;2.School of Information Engineering, Beijing Institute of Petro-chemical Technology,Beijing 102617,China)
Abstract:How to accurately locate the face is a very critical and important step in face recognition.This paper proposed a method of face detection based on eye location using OpenCV. Using the single frame image that was randomly got from the video images of the camera as the face image to be processed,a standardized face image can be received through eye location,image rotation,segmentation and scaling.Once more through the learning awareness of ART2 Neural Network,the rate of the recognition was calculated.The comparison experiments show that this method can be applied in face recognition and has relatively rapid speed of recognition and a good rate of recognition,which is particularly convenient in practical application.
Keywords:eye location  face detection  video image  neural network  face recognition
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号