Method of detection, division, and expression recognition of human face based on layered TDP model

Abstract

The invention discloses a method of detection, division, and expression recognition of the natural face based on a layered TDP model. Firstly, an original image is subjected to pre-treatment; an SIFT characteristic and corresponding position information are extracted; the two characteristics are combined by adopting the effective transference Dirichlet process to obtain a characteristic vector with geometric constraint; and the characteristic vector is input to the TDP model to obtain a first layer result-a human face or a non-human face. Then, the human face after division is taken as input of a second layer, and a division result of a subarea is obtained through the same process. Finally, the subarea is taken as the input of a third layer that is an expression recognition layer of the human face, and a result of expression recognition of a human face image is obtained by the same processes of characteristic extraction and combination. According to the invention, the problems that a model is needed to be established for every gesture in conventional multi-gesture expression recognition, and the model recognition rate is low due to factors such as a gesture are solved, and accuracy of expression recognition of a multi-gesture human face image is effectively improved.

Description

technical field [0001] The invention belongs to the field of emotion recognition, and in particular relates to a method for face detection, segmentation and expression recognition in a natural environment based on a hierarchical TDP model. Background technique [0002] Psychologist J.A. Russell proposed that in people's daily communication, only 7% of the information is transmitted through language, while 55% of the information is transmitted through facial expressions. It can be seen that facial expression is a very important way of information transmission. It is a rich source of information for human behavior and can convey information that many languages ​​cannot convey. In recent years, with the continuous improvement of some applications, the development of facial expression recognition technology has been promoted. [0003] In A.S.Willsky et al., entitled "Describing Visual Scenes Using Transformed Objects and Parts", the transfer Dirichlet process TDP is used to lea...

AI Technical Summary

Problems solved by technology

[0003] In A.S.Willsky et al., entitled "Describing Visual Scenes Using Transformed Objects and Parts", the transfer Dirichlet process TDP is used to learn the category of a region in a complex scene. This method combines geometric features and textures features to improve the recognition rate, but it can only be applied to scene images from specific angles, not to multi-angle natural face images

In addition, in the paper entitled "Facial Expression Recognition via a Boosted Deep Belief Network" by P. Liu et al., the independent feature extraction, feature selection and facial expression recognition processes in traditional learning are integrated into a unified It is completed in a cascaded deep neural network. This method improves the recognition rate of facial expressions by using the semantic information of the context, but the feature extraction is still based on a complete face, and the contribution to facial expression recognition is selected through feature selection. The largest feature, but due to the limitations of the feature selection method, it is impossible to guarantee whether the selected feature is the feature that can best meet the facial expression recognition

At present, there is no unified model that can efficiently and accurately recognize emotions in natural environments.

Images