基于Haar-Like T特征的人脸检测算法  

Face Detection Algorithm Based on Haar-Like T Features

作　　者：胡念 张四平[1] 王梅[1] Hu Nian;Zhang Siping;Wang Mei(School of Software,Hunan College of Information,Changsha Hunan 410200,China)

机构地区：[1]湖南信息职业技术学院软件学院,湖南长沙410200

出　　处：《信息与电脑》2020年第17期51-53,共3页Information & Computer

基　　金：长沙市社科规划一般项目“基于‘互联网+’推进智慧医疗建设研究”(项目编号:2020csskktzc63);湖南省教育厅科学研究项目“基于云计算人脸表情特征提取与识别的研究”(项目编号:XJK016CXX005)。

摘　　要：传统的人脸检测算法日渐成熟,但由于人脸五官的复杂性和外物的遮挡,使得人脸识别的准确率大大降低。在此背景下,笔者提出以Haar-Like T特征为基础的人脸检测算法,运用Adaboost分类器进行特征选择。实验结果表明:相比于传统的Haar-Like人脸检测算法,本文所提算法的检测率更高,误检个数更少,显著提升了人脸检测的效率。Traditional face detection algorithms are becoming more mature,due to the complexity of facial features and occlusion of foreign objects,the accuracy of face recognition is greatly reduced.This paper is thus motivated to propose a face detection algorithm based on Haar-Like T features in terms of Adaboost classifier for feature selection.Experimental results show that the proposed algorithm has a higher detection rate and a lower false detection rate,which significantly improves the efficiency of face detection.

关 键 词：人脸检测算法 HAAR-LIKE特征 Haar-Like T特征 ADABOOST分类器 

分 类 号：G710[文化科学—职业技术教育学]