Development of image processor "Tracking Vision" for sensing moving objects in an unstable illuminant environment.

Auteur(s)
Tomoto, S. & Sawasaki, N.
Jaar
Samenvatting

This article proposes a system to detect and track the running vehicle(s) under unstable light part of load (outside tunnel) at high accuracy. This system consists of the detecting part and tracking part. At the detecting part, optical flow observation points decided by locally correlated calculation are arranged at perpendicular to the vehicle running direction and in one row. The monitor detects vehicle(s) through the time dependent change of the moving vector in the running direction. By arranging the optical flow observation points in a row, volume of calculation was reduced. The vehicle tracking part carries out the template matching using locally correlated calculation. By selecting the template at the position having high texture characteristics, even if the shape of vehicle is flat and/or the direction of vehicle changes at curved road, high accuracy of tracking can be obtained. The ability of acquisition for running vehicles could be progressed to a practical level by this method. For the covering abstract see ITRD E114174.

Publicatie aanvragen

3 + 14 =
Los deze eenvoudige rekenoefening op en voer het resultaat in. Bijvoorbeeld: voor 1+3, voer 4 in.

Publicatie

Bibliotheeknummer
C 24446 (In: C 22454 CD-ROM) /73 / ITRD E115579
Uitgave

In: From vision to reality : proceedings of the 7th World Congress on Intelligent Transportation Systems ITS, Turin, Italy, 6-9 November 2000, 8 p., 3 ref.

Onze collectie

Deze publicatie behoort tot de overige publicaties die we naast de SWOV-publicaties in onze collectie hebben.