深度学习人脸识别算法在课堂考勤中的应用研究

Application of Deep Learning Face Recognition Algorithm in Class Attendance

人脸识别技术作为一种生物识别技术,已经广泛地应用于各个领域内的身份验证系统,课堂考勤系统是人脸识别技术应用的重要领域。传统的人脸识别课堂考勤存在着识别速度慢、准确率低的问题。随着深度学习的流行,基于深度学习的人脸识别逐渐取代了传统的人脸识别。本文利用了基于深度学习的SSD目标检测算法、FaceNet算法和SENet算法,结合谷歌的深度学习框架平台Tensor⁃Flow,对传统的人脸识别课堂考勤系统进行了改进和优化,有效地提升了人脸识别考勤的效率和准确度。

As a biometric technology,face recognition technology has been widely used in various fields of identity verification system,classroom attendance system is an important application field of face recognition technology.Traditional face recognition classroom attendance has the problems of slow recognition speed and low accuracy.With the popularity of deep learning,face recognition based on deep learning gradually replaces the traditional face recognition.This paper uses SSD target detection algorithm based on deep learning,facenet algorithm and senet algorithm,combined with Google’s deep learning framework platform tensorflow,to improve and optimize the traditional face recognition class attendance system,effectively improving the efficiency and accuracy of face recognition attendance.