Abstract
In this paper we present two low cost, airborne sensor systems capable of long-term vehicle tracking. Based on the properties of the sensors, a method for automatic real-time, long-term tracking of individual vehicles is presented. This combines the detection and tracking of the vehicle in low frame rate image sequences and applies the lagged Cell Transmission Model (CTM) to handle longer tracking outages occurring in complex traffic situations, e.g. tunnels. The CTM model uses the traffic conditions in the proximities of the target vehicle and estimates its motion to predict the position where it reappears. <br><br> The method is validated on an airborne image sequence acquired from a helicopter. Several reference vehicles are tracked within a range of <i>500m</i> in a complex urban traffic situation. An artificial tracking outage of <i>240m</i> is simulated, which is handled by the CTM. For this, all the vehicles in the close proximity are automatically detected and tracked to estimate the basic density-flow relations of the CTM model. Finally, the real and simulated trajectories of the reference vehicles in the outage are compared showing good correspondence also in congested traffic situations.
Highlights
The automatic real-time long-term tracking of vehicles using airborne optical camera systems can be relevant in a wide range of applications, e.g. transport monitoring, car fleet management, VIP vehicle monitoring during mass events or automotive industry applications
Two in-house developed, real-time optical camera systems were developed for the automatic tracking of specific vehicles, the CHICAGO system on a motorized glider (Runge, 2012) and the 4k system on a helicopter (Kurz, 2014)
The cell-transmission model Cell Transmission Model (CTM) (Daganzo, 1994) can be used to predict trajectories in complex traffic situations for a specific vehicle in occluded areas, such as tunnels or roads under bridges, or for occlusions caused by buildings and vegetation
Summary
The automatic real-time long-term tracking of vehicles using airborne optical camera systems can be relevant in a wide range of applications, e.g. transport monitoring, car fleet management, VIP vehicle monitoring during mass events or automotive industry applications. In the second part of this paper, a method for automatic real-time long-term tracking of a specific vehicle is presented, which combines the detection and tracking of vehicles in low frame rate image sequences (Szottka, 2011; Leitloff, 2014) and the lagged cell transmission model CTM (Daganzo, 1999) to bypass longer tracking outages in complex traffic situations. This method is demonstrated on an airborne dataset acquired on 16th October 2014 over Munich/Germany
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
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 callMore From: ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences
Disclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
