Application Research Of Face Emotion Identification Based On Deep Learning In Wisdom Education

Today,in today's rapid development,the state has also paid attention to the cultivation of talents.At the same time,the number of high-level teachers grows slowly,this situation leads to 1 teacher in the classroom to face at least 50 students,and only 1 teacher can also pay attention to all student learning status.At the same time,the surveillance resources in the classroom have not been fully utilized.Its feature is only to help students find lost items,monitoring,etc.,a large number of monitoring resources are ignored.With the arrival of the big data era and the continuous development of deep learning,the emergence of various networks and network models,and the continuous development of smart education is also promoted while large data is reasonably utilized.This article will be identified by FASTER R-CNN network,by modifying the backbone network model and loss function parameters in the network,the best network is selected through comparative experiments,and it is applied to the field of smart education,which is reached The teacher understands the state of learning,and reasonably uses monitoring resources,and launched a learning analysis after the teacher's class and improve the teaching method to lay a solid foundation.The main research contents of this article are as follows:1.This thesis selects a data set with different characteristics,and is classified according to the student learning status.It is labeled for the data set.2.Using the Faster R-CNN network,training and draw a marked data set,and then use the training-welled model to detect and position the five senses of the classroom middle school students,and classify and identify the emotions reflected in the five officials.And compare the accuracy in different models and loss function parameters.3.Studying the application scenario in the field of smart education,after the results of experimentation,select the highest accuracy of the Internet to the classroom teaching,prove the feasibility of facial emotion recognition technology in the field of smart education in the form of a software system.