Paper Abstract and Keywords |
Presentation |
2013-01-25 11:05
Dirichlet distribution Particle Filter for Posterior probability Smoothing
-- Application for NBI Videoendoscopy -- Tsubasa Hirakawa, Toru Tamaki, Bisser Raytchev, Kazufumi Kaneda, Tetsushi Koide, Yoko Kominami, Rie Miyaki, Taiji Matsuo, Shigeto Yoshida, Shinji Tanaka (Hiroshima Univ.) MI2012-100 |
Abstract |
(in Japanese) |
(See Japanese page) |
(in English) |
In this paper, we propose the smoothing method for sequential data using Particle Filter, and apply to NBI video endoscopy. For early detection of colorectal cancer, the recognition technique using Bag--of--Features and SVM is proposed, and extended to NBI video. However, there is a problem that recognition results isn't stabilized because it recognizes independently with each frame. Therefore, we propose the smoothing method for sequential data using posterior probability of SVM with each frame to likelihood of Particle Filter. Experimental results using NBI video shows that we can get more stable results. |
Keyword |
(in Japanese) |
(See Japanese page) |
(in English) |
NBI videoendoscopy / SVM / Bag--of--Visual Words / Dirichlet distribution / Particle Filter / / / |
Reference Info. |
IEICE Tech. Rep., vol. 112, no. 411, MI2012-100, pp. 201-206, Jan. 2013. |
Paper # |
MI2012-100 |
Date of Issue |
2013-01-17 (MI) |
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 |
MI2012-100 |
|