摘要
由于人的情感包含大量的信息,特别是面部的表情与手势图像中存在大量干扰信息,造成了情感计算正确率下降,为此提出了一种从用户的面部表情与手势中系统地分析情感线索的方法。对面部表情与手势进行分析是感情丰富的人机交互系统的基本组成部分,采用非语言线索算法来判断用户的感情状态。从图像序列中提取与表情相关的特征,通过智能法则系统分析用户的情感状态,得到的感情信息最终与用户的真实反映相适应。最后采用基于主体接口技术来处理如沮丧或愤怒等特定的情绪状态。实验结果表明,提出的方法能够准确地分析计算出用户的情感信息。
Especially facial expression and gesture image existed in a large number of interference information disadvantages,resulting in affective computing accuracy down,this paper presented a from the user’s facial expressions and gestures in a systematic analysis of the emotional cues of method.First analysis of facial expressions and gestures were the feelings of the rich interactive basic part of the system,and then the nonverbal cued algorithm to judge user emotional state.At the same time,these papers extracted from an image sequence and expression of related characteristics,through the intelligent rule system analysis of the user’s emotional state,and reflected the user’s suit.Finally based on the main interface technology,it treated such as frustration or anger and other specific emotional states.The experimental results show that,the proposed method can accurately analyze and calculate the user’s emotional information.
出处
《计算机应用研究》
CSCD
北大核心
2012年第7期2763-2765,2769,共4页
Application Research of Computers
基金
湖南省自然科学基金资助项目(11JJ6056)
关键词
情感计算
表情识别
人机交互
图像序列
affective computing
facial expression recognition
human-computer interaction
image sequence
