摘要
本文提出一种基于卷积神经网络CNN人脸识别模型,并将该模型应用于高职院校学生课堂行为分析。实验证明,使用卷积神经网络深度学习框架提取人脸深度特征,构建深度学习人脸识别模型,完成人脸识别,相比传统的人工设计的人脸特征提取,大大提高人脸识别的准确率。学生课堂行为识别算法可以正确判断学生的课堂行为,为课堂教学评价提供依据,实现更有效地教学,切实提高教学质量。
There are many studies on the face recognition model based on Convolutional Neural Network( CNN),and the technology is very mature. However,few scholars have applied CNN face recognition model to the study of college students ’classroom behavior analysis. Therefore,this paper proposes a CNN face recognition model based on Convolution Neural Network,and applies this model to the classroom behavior analysis of students in higher vocational colleges. Experiments show that the Convolutional Neural Network deep learning framework is used to extract face depth features,construct a deep learning face recognition model,and complete face recognition. Compared with the traditional manual design of face feature extraction,it greatly improves the accuracy of face recognition. The recognition algorithm of students’ classroom behavior can correctly judge students’ classroom behavior,which could provide a basis for classroom teaching evaluation,achieve more effective teaching and effectively improve teaching quality.
作者
左国才
吴小平
苏秀芝
王海东
ZUO Guocai;WU Xiaoping;SU Xiuzhi;WANG Haidong(Hunan Vocational Institute of Software,Xiangtan Hunan 411100,China;Hunan University,Changsha 410082,China)
出处
《智能计算机与应用》
2019年第6期107-110,共4页
Intelligent Computer and Applications
基金
湖南省教育科学规划课题研究成果(XJK19CZY018)
关键词
卷积神经网络
人脸识别模型
课堂行为分析
Convolutional Neural Network
face recognition model
classroom behavior analysis
