Facial expression recognition method, device and apparatus and storage medium

Abstract

The invention provides a facial expression recognition method, device and apparatus, and a storage medium. The method comprises the following steps of selecting a training data set and a to-be-recognized data set from a facial expression image; constructing a facial expression recognition model based on a residual neural network; selecting a part of facial expression images from the training dataset, inputting the facial expression images into a facial expression recognition model, and performing iterative training to obtain a trained facial expression recognition model; and selecting the part of facial expression images from the to-be-recognized data set, inputting the facial expression images into the trained facial expression recognition model, and performing facial expression recognition to obtain a facial expression classification result. The facial expressions can be accurately recognized and classified, the trained residual neural network model is used for real-time predictionunder the complex actual condition, the calculation time can be shortened, and facial expression recognition is rapidly achieved.

Description

technical field [0001] The present invention relates to the technical field of image processing, in particular to a facial expression recognition method, device, equipment and storage medium. Background technique [0002] In recent years, with the development of science, digital graphic information processing has begun to gradually explore in the direction of image emotion recognition and calculation. With the rise of artificial intelligence, research hotspots in the field of human-computer interaction have gradually increased. Facial expression recognition, which is related to computer science, biology, psychology and other disciplines, has also become one of the research hotspots. It is a direction with research prospects and research value. Its application can be extended to public opinion analysis, man-machine Interaction, medical treatment, fatigue driving supervision and other fields. For example, in the field of smart packaging, the facial expressions of customers a...

AI Technical Summary

Problems solved by technology

[0004] In view of the problems of high computational cost, long training time and low recognition accuracy of the traditional complex neural network face recognition system, the present invention provides a face expression recognition method based on the residual neural network, in an end-to-end manner, in advance After training the residual neural network model, using optimization techniques such as downsampling and pooling to reduce parameters and simplify the model structure, an expression classification model with high accuracy is obtained

Images