计算机与现代化
計算機與現代化
계산궤여현대화
COMPUTER AND MODERNIZATION
2015年
5期
1-8,12
,共9页
Kinect%行为识别%融合特征%SVM%HMM
Kinect%行為識彆%融閤特徵%SVM%HMM
Kinect%행위식별%융합특정%SVM%HMM
Kinect%activity recognition%fusion features%SVM%HMM
人体行为识别对于个人辅助机器人和智能家居等一些智能应用，是非常必要的功能，本文运用SVM＆HMM混合分类模型进行日常生活环境的人体行为识别。首先，使用微软的Kinect（一种RGBD感应器）作为输入感应器，提取融合特征集，包括运动特征、身体结构特征、极坐标特征。其次，提出SVM＆HMM模型， SVM＆HMM二级模型发挥了SVM和HMM各自的优点，既结合了SVM适于反映样本间差异性特点，又发挥了HMM适合处理连续行为的特点。该二级模型克服了单一SVM模型、传统HMM模型和在人体复杂和相似行为建模过程中精度、鲁棒性和计算效率上的不足。通过大量实验，结果表明SVM＆HMM二级模型对室内日常行为的识别具有较高的识别率，且具有较好的区分性和鲁棒性。
人體行為識彆對于箇人輔助機器人和智能傢居等一些智能應用，是非常必要的功能，本文運用SVM＆HMM混閤分類模型進行日常生活環境的人體行為識彆。首先，使用微軟的Kinect（一種RGBD感應器）作為輸入感應器，提取融閤特徵集，包括運動特徵、身體結構特徵、極坐標特徵。其次，提齣SVM＆HMM模型， SVM＆HMM二級模型髮揮瞭SVM和HMM各自的優點，既結閤瞭SVM適于反映樣本間差異性特點，又髮揮瞭HMM適閤處理連續行為的特點。該二級模型剋服瞭單一SVM模型、傳統HMM模型和在人體複雜和相似行為建模過程中精度、魯棒性和計算效率上的不足。通過大量實驗，結果錶明SVM＆HMM二級模型對室內日常行為的識彆具有較高的識彆率，且具有較好的區分性和魯棒性。
인체행위식별대우개인보조궤기인화지능가거등일사지능응용，시비상필요적공능，본문운용SVM＆HMM혼합분류모형진행일상생활배경적인체행위식별。수선，사용미연적Kinect（일충RGBD감응기）작위수입감응기，제취융합특정집，포괄운동특정、신체결구특정、겁좌표특정。기차，제출SVM＆HMM모형， SVM＆HMM이급모형발휘료SVM화HMM각자적우점，기결합료SVM괄우반영양본간차이성특점，우발휘료HMM괄합처리련속행위적특점。해이급모형극복료단일SVM모형、전통HMM모형화재인체복잡화상사행위건모과정중정도、로봉성화계산효솔상적불족。통과대량실험，결과표명SVM＆HMM이급모형대실내일상행위적식별구유교고적식별솔，차구유교호적구분성화로봉성。
Absrt act:Being able to recognize human activities is essential for several intelligent applications , including personal assistive ro-botics and smart homes .In this paper , we perform the recognition of the human activity based on the combined SVM&HMM in daily living environments .Firstly, we use a RGBD sensor ( Microsoft Kinect ) as the input sensor , and extract a set of the fusion features, including motion, body structure features and joint polar coordinates features .Secondly, we propose a combined SVM&HMM Model which not only combines the SVM characteristics of reflecting the difference among the samples , but also de-velops the HMM characteristics of dealing with the continuous activities .The SVM&HMM model plays their respective advantages of SVM and HMM comprehensively .Thus, the combined model overcomes the drawbacks of accuracy , robustness and computa-tional efficiency compared with the separate SVM model or the traditional HMM model in the human activity recognition .The ex-periment results show that the proposed algorithm possesses the better robustness and distinction .
