Expression recognition model construction method and system

Abstract

The invention relates to an expression recognition model construction method and system, and relates to the field of data processing. The method comprises: S1, acquiring and preprocessing a source image to obtain a preprocessed image; S2, constructing a deep learning model according to the source image and the preprocessed image, and introducing transfer learning to carry out weight training on the deep learning model to obtain an image pre-training model; S3, obtaining a fine adjustment model according to the image pre-training model; and S4, performing fine adjustment model training on the preprocessed image by using the fine adjustment model to obtain an expression recognition model. According to the scheme, the technical problem of how to improve the accuracy of the expression recognition model is solved, and the method is suitable for expression recognition.

Description

technical field [0001] The invention relates to the field of data processing, in particular to a method and system for constructing an expression recognition model. Background technique [0002] At present, the construction methods of facial expression recognition models are mainly divided into two categories: one is based on the facial action coding system, which detects facial muscle movements and constructs the mapping relationship between them and emotions to achieve the purpose of expression recognition; It adopts the method of deep learning, through the end-to-end learning method, automatically extracts the characteristics of human expressions, and performs expression recognition. [0003] At present, the facial expression recognition model construction method is difficult to achieve a high accuracy rate. The main problems are as follows: the accuracy of expression recognition based on deep learning often relies on a huge training data set, and the expression data set ...

AI Technical Summary

Problems solved by technology

[0003] At present, the facial expression recognition model construction method is difficult to achieve a high accuracy rate. The main problems are as follows: the accuracy of expression recognition based on deep learning often relies on a huge training data set, and the expression data set is difficult to construct and obtain on a large scale. Training requirements, at the same time, lighting factors, obstacle occlusion or face posture are all interference factors that affect facial expression recognition

Images