Presentation 2001/3/15
Front Scene Observation of Vehicle using Energy Function for Correspondence Estimation
Sadaatsu KATO, Kazuyuki SASAKI, Masato NAKAJIMA,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) In this paper we propose a new technique for front scene observation from the cruising vehicle to prevent the collision accident using sequence of road images. This method rapidly calculates optical-flow with minimization of Energy Function all over the image pixels. At first, all image pixels are supposed that they exist on road plane and calculated initial value of value of optical flow. Then comparison between calculated value of optical flow using the Energy Function and initial value of optical flow abstracts only optical flow emerging from front-vehicles. This technique was applied to the images acquired on the expressway, and the result suggests that this approach can effectively detect front vehicles and give alarms to drivers.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) road traffic / image processing / front scene observation / optical flow / energy function
Paper # ISEC2000-108
Date of Issue

Conference Information
Committee ISEC
Conference Date 2001/3/15(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Information Security (ISEC)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Front Scene Observation of Vehicle using Energy Function for Correspondence Estimation
Sub Title (in English)
Keyword(1) road traffic
Keyword(2) image processing
Keyword(3) front scene observation
Keyword(4) optical flow
Keyword(5) energy function
1st Author's Name Sadaatsu KATO
1st Author's Affiliation Keio UniVersity()
2nd Author's Name Kazuyuki SASAKI
2nd Author's Affiliation YAZAKI CORPORATION
3rd Author's Name Masato NAKAJIMA
3rd Author's Affiliation Keio UniVersity
Date 2001/3/15
Paper # ISEC2000-108
Volume (vol) vol.100
Number (no) 691
Page pp.pp.-
#Pages 6
Date of Issue