DocumentCode :
2627751
Title :
Extended CoHOG and particle filter by improved motion model for pedestrian active safety
Author :
Kataoka, Hirokatsu ; Tamura, Kimimasa ; Aoki, Yoshimitsu ; Matsui, Yasuhiro
Author_Institution :
Grad. Sch. of Sci. & Technol., Keio Univ., Kanagawa, Japan
fYear :
2012
fDate :
25-28 Oct. 2012
Firstpage :
1511
Lastpage :
1516
Abstract :
The percentage of pedestrian deaths in traffic accidents is on the rise. In recent years, there have been calls for measures to be introduced to protect such vulnerable road users as pedestrians and cyclists. In this study, a method to detect pedestrians using an in-vehicle camera is presented. We improved the technology in detecting pedestrians with highly accurate images using a monocular camera. We were able to predict pedestrians´ activities by monitoring them, and we developed an algorithm to recognize pedestrians and their movements more accurately. The effectiveness of the algorithm was tested using images taken on real roads. For the feature descriptor, we found that an extended co-occurrence histogram of oriented gradients, accumulating the integration of gradient intensities. In tracking step, we applied effective motion model using optical flow for Particle Filter tracking. These techniques are valified by using images captured on the real road.
Keywords :
cameras; gradient methods; image motion analysis; image sequences; object recognition; object tracking; particle filtering (numerical methods); pedestrians; road safety; road traffic; statistical analysis; traffic engineering computing; cooccurrence of histograms of oriented gradients; cyclists; extended CoHOG; feature descriptor; gradient intensity; in-vehicle camera; monocular camera; motion model; optical flow; particle filter tracking; pedestrian active safety; pedestrian activity prediction; pedestrian death; pedestrian detection; pedestrian recognition; traffic accident; Image edge detection; Shape;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
IECON 2012 - 38th Annual Conference on IEEE Industrial Electronics Society
Conference_Location :
Montreal, QC
ISSN :
1553-572X
Print_ISBN :
978-1-4673-2419-9
Electronic_ISBN :
1553-572X
Type :
conf
DOI :
10.1109/IECON.2012.6388518
Filename :
6388518
Link To Document :
بازگشت