[Skip to content]

World Highways ITS International
Search our Site
Siemens Web bannerTattile Logo 295 x 110Jenoptik Logo 295 x 110Dynniq logo 295 x 110 bannerIterisLogo2018_Black_avatar.jpgKapsch 2018_295x110.jpg
10/02/2012

New standards in automatic vehicle classification

[ Zoom ]
Telegra AVC
Telegra comes to Intertraffic to demonstrate why its Scanway automatic vehicle classification (AVC) system is setting new standards for accuracy, reliability, adaptability and robustness. The system has now been employed worldwide in various modern ITS infrastructures including automated toll collection, and traffic control in both highway and urban situations.

In claiming that its AVC sets new higher standards, Telegra points to Scanway’s precision in recognising over 50 types of vehicles, which can be grouped into over 20 vehicle classes. High quality of raw information acquired by IR vertical scanner allows precise extraction of a large number of vehicle features from the acquired vehicle profile. By examining these features the classification logic is able to determine the vehicle category. The obtained category can then be mapped into a user-defined vehicle class. According to the company, this principle provides high flexibility of the classification system, since vehicle classes can be adjusted practically on-the-fly, without modifying core classification parameters. High robustness of Telegra's AVC is achieved through redundant IR light grid, which allows error detection on each IR channel.

As an example of a real world deployment of its AVC system that backs up its claims, Telegra points to India where 180 systems have been successfully deployed to assist in toll collection on highways. In this particular case, thirty one categories are used that are mapped into one of ten classes defined to match the specific client’s requirements. The overall classification accuracy observed in real traffic conditions is 99.7%.

Stand: 11.201

Vitronic banner 295 x 1103M 295 x 110.jpgLindsay 295 x 110Flir 295 x 110 Swarco LogoADH1_Red_1797_295x110.jpg