Quickly Modeling Living Environment Based on Virtual Sensorization
A System for registering target behaviors quickly and measuring roboustly is requested, when we collect and analyze human behaviors in office, welfare institution and living environment. Conventional systems for measuring behaviors have some problems. For example, a measuring function is fragile using visual information, it is need to create complex programing for extract necessary information from information of several sensors which is embeded in environment.
This paper describes a method for robustly detecting and efficiently recognizing daily human behavior in real world. The proposed method is performed with the following steps: 1) real world sensorization for robustly observing his or her behavior using ultrasonic 3D tags, which is a kind of an ultrasonic location system, 2) real world virtualization for creating a virtual environment through modeling 3D shape of real objects by a stereovision system, and 3) virtual sensorization of the virtualized objects for quickly registering human behavior handling objects in real world and efficiently recognizing target human behavior.
Detection of Human Behavior modeling objects' function
- Enhancement of virtual sensor for increasing the number of function.
- Detection of longer-term behaviors, i.e. having a meal, which is collective of short-term behaviors.