基于模糊概念的人脸表情知识研究

Research on Knowledge Representation of Facial Expression Features Based on Fuzzy Concept

提出了一种基于人脸语义知识的表情描述方法。利用模糊集理论将表情特征转化为模糊概念,建立表情特征与模糊语义之间的映射关系,构建优化准则提取能够有效刻画不同表情之间特征的模糊概念,并利用模糊概念构建表情特征分类器。实验分别使用,FEI和CK+表情库进行分析,并与C4.5、Ripper、Decision Table以及Cart等方法进行比较。结果表明:文中构建的语义分类器与上述分类器具有相似的性能,且能够有效识别不同表情。

This paper proposes an expression description method based on face semantic knowledge.In the framework of axiomatic fuzzy sets,facial features are transformed into semantic concepts which are used as rule sets to distinguish expression categories.This method establishes the semantic description method of the underlying facial geometric features;changes the expression pattern of the fuzzy rule set;constructs optimization criteria,and extracts the semantic concepts to characterize typical features of facial expressions.In the experiments,the method of this paper was verified using FEI and CK+databases,and the results show that the method has good interpretability and classification performance for facial expressions.