A Novel Human Emotion Classification Method Based on Grid Environment

摘要：

Communication will only truly be possible if we are able to “see the other partys responses.Thus effective and fast Human Emotion Classification Method is essential.However,how to improve the performance and runtime of the human emotion classification is a challenging problem.This paper describes the use of statistical techniques and Hidden Markov Models(HMM)in the classification of emotions.The method aims to classify 6 basic emotions (angry,dislike,fear,happy,sad and surprise)from both facial expressions(video)and emotional speech (audio).In order to effectively improve the performance and reduce the runtime,the task is distributed in grid computing environment.The experimental results show that the performance in the accuracy of classification and the processing time are improved significantly compared to that of the traditional method.

关键词： grid computing human emotional classification hidden Markov model nearest neighbor algorithm

作者: Zhong Gao Daquan Gu Meng Cui

作者单位: Collegel of Telecommunications and Information Engineering,Nanjing University of Posts and Telecommu Institute of Meteorology,PLA University of Science and Technology,Nanjing,China

会议类型: 国际会议

会议名称: 2008亚太心智、脑和教育学术会议(Asia-Pacific Conference on Mind Brain and Education 2008)

会议地点: 南京

会议语种:英文

页码: 160-163

在线出版日期: 2008-10-25（万方平台首次上网日期，不代表论文的发表时间）