Paper Abstract and Keywords |
Presentation |
2008-12-18 14:50
Activity Recognition from Interactions with Objects using Dynamic Bayesian Network Tomohito Inomata (ATR Knowledge Science Laboratories), Futoshi Naya (Ritsumeikan Univ.), Noriaki Kuwahara (Graduate School of Science and Tech/Kyoto Inst. of Tech.), Fumio Hattori (Graduate Scvhool of Sciemce and Engineering/Ritsumeikan Univ.), Kiyoshi Kogure (ATR Knowledge Science Laboratories) PRMU2008-163 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
A nursing activity recognition method has been developed that is based on a nurse's interactions with the tools and materials he/she touches and uses the Dynamic Bayesian Network(DBN) framework. This is aimed to prevent the cause of medical accident and incidents beforehand. This time, we focused on recognizing the 12 activity steps in the drip injection task which is one of the most related tasks to medical accidents and incidents. Activity recognition from the interactions with objects captured by RFID was done using DBN framework, and we obtained the 96% accuracy of the recognition of the 12 activity steps in the drip injection task. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
RFID / DBN / Activity Recognition / / / / / |
Reference Info. |
IEICE Tech. Rep., vol. 108, no. 363, PRMU2008-163, pp. 93-98, Dec. 2008. |
Paper # |
PRMU2008-163 |
Date of Issue |
2008-12-11 (PRMU) |
ISSN |
Print edition: ISSN 0913-5685 Online edition: ISSN 2432-6380 |
Copyright and reproduction |
All rights are reserved and no part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopy, recording, or any information storage and retrieval system, without permission in writing from the publisher. Notwithstanding, instructors are permitted to photocopy isolated articles for noncommercial classroom use without fee. (License No.: 10GA0019/12GB0052/13GB0056/17GB0034/18GB0034) |
Download PDF |
PRMU2008-163 |
|