Key points are not available for this paper at this time.
In this paper we are presenting our take on object recognition in traffic monitoring systems. In the cooperation of university with company Betamont we aim to enlarge the portfolio of products for traffic monitoring with new approaches and improved technologies. Ability to reliable detect moving vehicles and correctly recognize vehicle type (car, bus, truck) make (brand) and model allows for non-invasive monitoring methods (without the use of road sensors) to be used and for additional important information about traffic to be gathered. And as such, these additional data can be utilized in the projects of smart cities and smart traffic control systems in the future. In this paper, we are focusing on the description of our current stage of research, we provide results to preliminary test done with system using local feature extraction and MF ArtMap classification on real data from monitoring point in Banská Bystrica, and we discuss its benefits and shortcomings.
Lorenčík et al. (Wed,) studied this question.