基于多Agent的情感ICAI系统的研究  

RESEARCH ON EMOTIONAL ICAI SYSTEM BASED ON MULTI-AGENT

作　　者：谢文秀[1] 段富[1] 

机构地区：[1]太原理工大学计算机科学与技术学院,山西太原030024

出　　处：《计算机应用与软件》2014年第1期65-68,共4页Computer Applications and Software

基　　金：山西省自然科学基金项目(2008011039)

摘　　要：提出一种可以识别和理解学习者的情感,并能针对学习者情感做出反应的智能计算机辅助教学系统ICAI(intelligent computer assisted instruction system)。系统结合表情识别和多Agent技术,依据人脸图像的分割信息,提出新的表情特征提取的方法,分别对眼睛、眉毛、嘴巴进行特征提取,将其应用于表情识别模块获得了很好的效果;同时,针对系统功能,引入多Agent技术实现各模块的主要功能。实验结果表明,该系统可以有效解决网络教学中情感缺乏的问题,具有良好的执行效率。We present such an ICAI teaching system which can recognise and understand learner＇ s emotion and can make response to learners＇ emotion. The system combines the facial expression recognition and the multi-agent technology, according to the segmentation information of face image, a new expression feature extraction method is proposed and is used to carry out the features extraction on eyes, eyebrows and mouth respectively, and use them to emotion recognition module, the effect is very good. At the same time, in view of the system function, we introduce multi-agent technology to realise the main functions of each module. Demonstrated by the results of the experiment, the system can effectively solve the oroblem of the lack of emotion in network tenehint~ ~v^t~rn nnrl hn~ ,~,~nrl ,~,~o,,t;,,,~ ,~;,-;

关 键 词：情感 ICAI 特征提取 AGENT 

分 类 号：TP396.1[自动化与计算机技术—计算机应用技术]