Research On Learning Interest Evaluation Method Based On Facial Expressions

Facial expressions contain rich inner emotional information and are one of the important ways for people to communicate emotionally.Facial expression recognition in the learning state is the key to the intelligent teaching system.Timely recognition of facial expression in the learning state is helpful for teachers or parents to judge whether learners are interested in the learning content,so as to take timely adjustment measures and improve learning efficiency.Deep learning end-to-end model was used to the raw data as input,the final recognition results as output,the feature extraction and feature classification combination,avoid the traditional methods of multifarious artificial,feature extraction and classification process and deep learning approach is to rely on data driven,such as light,gesture,shelter problem not sensitive,widely used in the field of machine learning and artificial intelligence.In order to improve the accuracy and generalization ability of facial expression recognition,this paper adopts the convolutional neural network method in deep learning to carry out related research on facial expression recognition.The main work is as follows:1.In order to improve the accuracy of facial expression recognition,an improved method of convolutional neural network is proposed,which adopts the method of small size convolution kernel and continuous convolution layer to extract more detailed local features and increase the nonlinear expression ability of the network.The experiment proves that this method can effectively enhance the feature extraction ability of the model.2.In order to further improve the accuracy and efficiency of facial expression recognition,batch normalization layer was added after each convolutional layer,and the effectiveness of this method was verified by experiments on Fer2013 facial expression data set.3.The most common expressions of sleepiness and concentration in the learning process were introduced to expand the database of expressions and establish the learning database of expressions.4.A learning interest detection system based on facial expression recognition is built,which can realize face detection,learning facial expression recognition classification and other functions.