For Full-Text PDF, please login, if you are a member of IEICE,|
or go to Pay Per View on menu list, if you are a nonmember of IEICE.
Robust Background Subtraction Based on Statistical Reach Feature Method
Kenji IWATA Yutaka SATOH Ryushi OZAKI Katsuhiko SAKAUE
D - Abstracts of IEICE TRANSACTIONS on Information and Systems (Japanese Edition)
Publication Date: 2009/08/01
Online ISSN: 1881-0225
Print ISSN: 1880-4535
Type of Manuscript: Special Section PAPER (Special Section on Image Recognition and Understanding)
background subtraction, statictical reach feature, SRF, ISC, RRC, feature extraction,
Full Text(in Japanese): PDF(800.3KB)
>>Buy this Article
We propose a statistical reach feature (SRF) method that extracts the robust features for noises, and a background subtraction method based on the SRF method that is robust for spatial and sequential noises such as illumination changes and/or waving trees. Extracting robust features in the given image sequence is an important fundamental technology which influences the performance of various computer vision systems. The SRF method selects stable pairs of coordinates that keep the sign of increase and decrease of the brightness in the image sequence. This paper describes a definition of the SRF method, and the application to background subtraction. The experimental results show that the proposal method can segment the background with stability even if the image sequence contains extreme illumination changes.