人脸面部混合表情识别系统

The Human Facial Combined Expression Recognition System

根据心理学家对表情的研究和前人的工作成果 ,在对动态表情图像序列的时序分析的基础上 ,提出了对混合表情的识别系统 .把脸部分成各个表情特征区域 ,分别提取其运动特征 ,按时序组成特征序列 ,通过分析不同特征区域所包含的不同表情信息的含义和表情的含量 ,识别任意时序长度的。

The human beings have not only rational thinking and logic reasoning ability, but also perceptual thinking and emotions. In the aim of the natural and intelligent human machine communication, it is essential to make it possible that the computer can understand and express the emotions. This paper summarizes the study in the psychology field and the former work and proposes the combined expression recognition system based on the analysis of the dynamic expression image sequences. The face is taken as being composed of several primary expression regions, in which the motion features can be extracted and constituted to eigen sequences. The analysis of the arbitrary length of image sequences of facial expressions and combined expression recognition are proposed and implemented by analyzing the respective expression meaning and the expression contents of different primary regions and using the multi feature fusion.