基于傅里叶-隐马尔科夫模型的人体行为识别  

Human Behavior Recognition Based on HMM

作　　者：黄静[1] 孔令富[1] 李海涛[1] 

机构地区：[1]燕山大学信息学院,河北秦皇岛066004

出　　处：《计算机仿真》2011年第7期245-248,共4页Computer Simulation

基　　金：河北省自然科学基金资助项目(F2010001295);国家自然科学基金项目(60975062)

摘　　要：研究人行为识别的正确性,针对提高对人体行为序列图像进行识别的能力,隐马尔科夫模型(HMM)是一种统计分析模型,具有时序模式分析能力。为了增加图像信息的有效性,提出了一种傅里叶与隐马尔科夫模型相结合的方法人体行为识别方法。通过获得各种人体行为的二值图像序列,对待识别的序列图像提取具有旋转、平移和尺度不变性的傅立叶特征,采用了一种基于中心距的傅里叶描述子,利用改进的隐马尔科夫模型对提取的特征向量进行分类,得到人体行为的识别结果。试验结果表明,系统的识别率与HMM的状态数和观察值数有关,方法是有效且可行的,设计适当的HMM分类器能使系统的识别率达到90%以上,实际应用效果满足要求。Hidden Markov Model is a statistical analysis model,with great abilities to analyse time variant patterns.In order to increase the effectiveness of image information and improve the capability of recognizing the image sequences of human behaviors,a novel method based on the combination of Fourier descriptor and Hidden Markov Model is proposed to recognize the serial image of human actions.The paper first obtains the binary image sequences,extracts the Fourier features with the characteristics of rotation,translation and scale invariances,and proposes a kind of Fourier transform based on the center distance,then uses the improved Hidden Markov Model to classify the extracted feature vectors and gets the recognition results.The experiment results show that the method proposed in this paper is effective and feasible,and the recognition rate is related to the state number and observation number of HMM,and the appropriate HMM classifier can make the recognition rate of the system rise be above 90%.

关 键 词：特征提取 傅立叶描述子 隐马尔科夫模型 人体行为识别 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]