Analysis of traffic data plays an important role in urban and spatialplanning. Thermal Infrared (TIR) video cameras have capabilitiesto operate at day and night and to acquire the scene sampled withvideo frame rate. In this paper a strategy for the estimation ofvehicle motion and the assessment of traffic activity from airborneTIR video is presented. In contrast to other approaches we handledetecting and tracking vehicles in the video separately, becausemoving as well as stationary vehicles are intended to be detected.Firstly, vehicles are detected in single frames of the video. Additionally,tie points are detected for co-registration and compensation thesensor movement. Afterwards, a stepwise grouping of image pointsconsidering temporal consistence and geometric relation is carriedout to determine the vehicle trajectories and classify them intostationary, moving and uncertain dynamical categories. The vehiclesare then integrated into the classes "moving," "stationary" and "uncertain"categories. Additionally, in consideration of matching vehicle-relatedimage patches for moving vehicles, the topology of the trajectoriesare investigated and optimized in order to eliminate disturbancesand estimate velocities. The algorithms were tested with video sequenceof urban areas in nadir-view and oblique-view. The correctness ofthe results is achieved higher than 75% for both views.     «

Analysis of traffic data plays an important role in urban and spatialplanning. Thermal Infrared (TIR) video cameras have capabilitiesto operate at day and night and to acquire the scene sampled withvideo frame rate. In this paper a strategy for the estimation ofvehicle motion and the assessment of traffic activity from airborneTIR video is presented. In contrast to other approaches we handledetecting and tracking vehicles in the video separately, becausemoving as well as stationary vehicles are...     »