Traffic data collection and analysis is essential for designing roundabouts and improving safety at roundabouts. One of the main interested traffic data for roundabouts is often gap size. However, manual collection of gaps size by human operators is extremely laborious. In this work, we developed a system/tool to allow data mining of the recorded videos for automated extraction of gap size. The developed system has two main modules. The tracking module is used to derive raw data, which are vast amount of resulting outputs of vehicle trajectories for all tracked vehicles, which provide rich information for traffic data extraction as the trajectories give the position, size, shape and speed of the vehicle at each time moment. A post-processing data mining module is then used to automatically extract the interested traffic data. The extracted gap size has been compared to ground truth manual measurements and it is shown that an accuracy of almost 100% has been achieved. Compared to manual inspection, the proposed system/tool significantly saves time and cost.
|Original language||English (US)|
|Title of host publication||2017 IEEE 30th Canadian Conference on Electrical and Computer Engineering, CCECE 2017|
|Publisher||Institute of Electrical and Electronics Engineers Inc.|
|State||Published - Jun 12 2017|
|Event||30th IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2017 - Windsor, Canada|
Duration: Apr 30 2017 → May 3 2017
|Name||Canadian Conference on Electrical and Computer Engineering|
|Other||30th IEEE Canadian Conference on Electrical and Computer Engineering, CCECE 2017|
|Period||4/30/17 → 5/3/17|
Bibliographical notePublisher Copyright:
© 2017 IEEE.
Copyright 2017 Elsevier B.V., All rights reserved.