No abstract

No abstract

In this paper, we propose the human behavior detection and activity support environment Vivid Room. Behavior in Vivid Room is detected by numerous sensors built into the room, i.e., magnet sensors for doors and drawers, microswitches for chairs, and ID tags for personnel, and information is collected by a sensor server via an RF tag system and LAN. To recognize meaningful behavior, e.g., studying, eating, and resting, we use ID4-based learning system. We also developed activity support using sound and voice taking into account human behavior in the room. Experimental results confirmed the accuracy of behavior recognition and the quality of support.

No abstract

No abstract