一种基于OpenCV的人脸实时识别系统  

The Real-time Facial Recognition System Based on OpenCV

作　　者：李宏伟 陈冰 高倩 Li Hongwei;Chen Bing;Gao Qian(Luohe Institute of Technology,Henan University of Technology,Luohe,Henan 462000,China)

机构地区：[1]河南工业大学漯河工学院,河南漯河462000

出　　处：《漯河职业技术学院学报》2024年第6期22-25,共4页Journal of Luohe Vocational and Technical College

基　　金：河南省软科学研究计划项目“‘产教融合科教融汇’背景下企业主导的河南省先进制造业协同创新模式实践探索”(242400410478)。

摘　　要：人工智能的快速发展使得图像处理技术扮演着越来越重要的角色,其人脸识别算法的应用也越来越广泛。本文结合OpenCV计算机视觉库,首先介绍3种经典的通用人脸识别算法,即基于LBFH分类器人脸识别算法、基于特征脸EigenFace分类器人脸识别算法、基于鱼脸特征FisherFace分类器人脸识别算法;并在树莓派4B硬件的基础上,使用基于LBPH人脸识别方法,搭建能够快速进行追踪识别的人脸识别系统,为识别准确率要求不高的场景提供一种技术方案。With the rapid development of artificial intelligence,image processing technology plays a more and more important role,and its facial recognition algorithms are applied more and more widely.By combining with the Opencv computer vision library,the paper introduces three classic universal facial recognition algorithms based on the LBFH classifier,the EigenFace classifier and the FisherFace classifier respectively.On the basis of Raspberry Pi 4B hardware,a facial recognition system that can quickly track and recognize is built by using the LBPH facial recognition method to provide a certain technical solution for scenarios with low recognition accuracy.

关 键 词：人工智能 图像处理 人脸识别 LBFH分类器 树莓派4B 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]