基于深度学习的多人脸识别系统的设计实现  

Design of Multi-face Recognition System Based on Deep Learning

作　　者：郭琪 王志慧 范道尔吉 武慧娟[1,2] GUO Qi;WANG Zhihui;FAN Daoerji;WU Huijuan(College of Electronic Information Engineering,Inner Mongolia University,Hohhot 010021,China;College of Energy and Power,Inner Mongolia University of Technology,Hohhot 010321,China)

机构地区：[1]内蒙古大学电子信息工程学院,呼和浩特010021 [2]内蒙古工业大学能源与动力学院,呼和浩特010321

出　　处：《内蒙古农业大学学报（自然科学版）》2022年第2期86-92,共7页Journal of Inner Mongolia Agricultural University(Natural Science Edition)

基　　金：国家自然科学基金项目(61763034)“;智能视觉物联网中人车物的图像检索技术”预研项目。

摘　　要：随着人们对社会安全保障的需求越来越大,对人脸识别的性能要求也越来越高。但是目前的人脸识别算法在面对多人脸识别的挑战时,存在准确率低和速度过慢的缺陷。本文利用PyQt GUI设计实现了基于深度学习的多人脸识别系统,该系统搭载了应用于人流量大场景下的多人脸识别算法,该算法可以很好地兼顾了精度和速度的要求。整体系统轻巧,易于在移动设备上搭建,界面简洁清晰、功能强大,可以为用户提供多人脸识别的各种功能。As people’s demands for social security are increasing,the performance requirements for the face recognition are also getting higher and higher. However,the current face recognition algorithm has the defects of low accuracy and slow speed when facing the challenge of multi-face recognition. A multi-face recognition system based on the deep learning was designed and implemented in this paper by using the PyQt Graphical User Interface(GUI)technology. The system was equipped with a multi-face recognition algorithm that could be applied to the scenarios with large human flow throughput,which could well take into account the requirements of both speed and accuracy. The system was lightweight and easy to build on mobile devices,with a friendly and clear interface,and powerful functions. It could provide users with various functions of multi-face recognition.

关 键 词：深度学习 多人脸识别 GUI 

分 类 号：TP317.4[自动化与计算机技术—计算机软件与理论]