基于机器视觉的在线人脸识别方法研究  

Research on online face recognition method based on machine vision

作　　者：谢宇宸 XIE Yuchen(Fuzhou No.1 High School,Fuzhou 350100,China)

机构地区：[1]福州市第一中学,福州350100

出　　处：《计算机应用文摘》2024年第16期194-196,共3页Chinese Journal of Computer Application

摘　　要：福州市第一中学(以下简称“我校”)的闸机人脸识别系统存在识别失败和识别错误的情况,引发了道路拥堵问题。对此,文章深入分析了机器视觉领域中的度量学习研究进展,对多分类学习及度量学习的研究工作进行了全面综述,旨在通过深度学习策略提高我校闸机人脸识别系统的识别有效性,从而有效解决拥堵问题,进而为深度学习在人脸识别方面的研究提供有益的参考意见。The facial recognition system of the gate machines at Fuzhou No.1 High School(hereinafter referred to as“our school”)has encountered recognition failures and errors,leading to road congestion issues.In this regard,the article delves into the research progress of metric learning in the field of machine vision,and provides a comprehensive review of the research work on multi class learning and metric learning.The aim is to improve the recognition effectiveness of our school̓s gate face recognition system through deep learning strategies,thereby effectively improving congestion problems and providing useful reference opinions for the research of deep learning in face recognition.

关 键 词：人脸特征 机器视觉 深度神经网络 深度学习 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]