Title :
A Home Environment Posture and Behavior Recognition System
Author :
Guo, Ping ; Miao, Zhenjiang
Author_Institution :
Beijing Jiaotong Univ., Beijing
Abstract :
In this paper, we propose a home environment health care surveillance system. This system can recognize six static postures and four activities, and send out informing SMS messages when abnormal actions are detected. The system contains two parts: static posture recognition and human behavior analysis. We use a modified shape context method which we call the area-based shape context to recognize static postures and Hidden Markov Models are used to build motion models.
Keywords :
Markov processes; behavioural sciences computing; health care; video surveillance; SMS messages; behavior recognition system; hidden Markov models; home environment health care surveillance system; home environment posture; human behavior analysis; posture recognition; Biological system modeling; Context modeling; Feature extraction; Hidden Markov models; Humans; Image motion analysis; Medical services; Motion analysis; Shape; Surveillance;
Conference_Titel :
Convergence Information Technology, 2007. International Conference on
Conference_Location :
Gyeongju
Print_ISBN :
0-7695-3038-9
DOI :
10.1109/ICCIT.2007.35
