Abstract
Since global road traffic is steadily increasing, the need for intelligent traffic management and observation systems is becoming an important and critical aspect of modern traffic analysis. In this paper, we cover the development and evaluation of a traffic measurement system for tracking, counting and classifying different vehicle types based on real-time input data from ordinary highway cameras by using a hybrid approach including computer vision and machine learning techniques. Moreover, due to the relatively low framerate of such cameras, we also present a prediction model to estimate driving paths based on previous detections. We evaluate the proposed system with respect to different real-life road situations including highway-, toll station- and bridge-cameras and manage to keep the error rate of lost vehicles under 10%.
Highlights
To develop a suitable traffic management or surveillance system the current traffic on the road in a specific area has to be identified and measured respectively
To develop a camera-based traffic measurement system which is capable of tracking, counting and classification of different vehicle types, a computer system has to analyze video frames using common computer vision and machine learning techniques in a hybrid application
After establishing a background model each input frame is subtracted by the static background model
Summary
To develop a suitable traffic management or surveillance system the current traffic on the road in a specific area has to be identified and measured respectively. To develop a camera-based traffic measurement system which is capable of tracking, counting and classification of different vehicle types, a computer system has to analyze video frames using common computer vision and machine learning techniques in a hybrid application. Sci. 2020, 10, 6270 centralized computer system It allows changes between common traffic hotspots in seconds by switching the input stream to another camera. The success of such a system mainly depends on five identified factors following the approach of using remote cameras:. A prediction model is established to enrich the raw video data with additional information It tries to find associations between the current and past detected vehicles in order to derive driving paths. A final conclusion and remarks about the system are given in the last section in order to provide a base for discussion, further research and future system revisions
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
