2011 Fiscal Year Final Research Report
A Research to Recognize User' s status in the Desk Work for Ambient Information Environment
Project/Area Number |
22700123
|
Research Category |
Grant-in-Aid for Young Scientists (B)
|
Allocation Type | Single-year Grants |
Research Field |
Media informatics/Database
|
Research Institution | Osaka University |
Principal Investigator |
YUICHI Itoh 大阪大学, 大学院・情報科学研究科, 准教授 (40359857)
|
Project Period (FY) |
2010 – 2011
|
Keywords | ヒューマンインターフェイス / アンビエント情報環境 / ユーザ状況認識 |
Research Abstract |
We have focused on the realization of an ambient information environment capable of supporting users by displaying information they intend as measured by multiple sensors. In order to establish a methodology to implicitly acquire a user' s situation and status, we developed a special chair in our research equipped with several sensors for desk work and confirmed its validity by conducting experimental evaluations. In 2010, we developed a chair system with four weight sensors, one at each corner of the seat and, by investigating its performance, were able to show that it was sensitive enough to measure a user' s movement and posture. In 2011, we conducted experiments to evaluate the chair system. Our results showed that the system could distinguish eleven different postures for eight participants with an accuracy of 70%. These postures were regarded as primitive(e. g., crossing one' s legs, resting one' s chin on one' s palm, etc.). Our goal was to develop a system capable of recognizing high-level contexts(e. g., fatigue, concentration, etc.). We investigated the relationship of the primitive postures to high-level contexts and were able to achieve a 90% recognition rate with the high-level contexts. In addition, our data showed that it took 10 seconds for participants' postures to stabilize after the participant started to sit, and that the recognition rate could be improved by taking this factor into consideration
|
Research Products
(7 results)