Estimating urban activity on high-resolution thermal image sequences aided by large scale vector maps

In this contribution we describe a method to assess the activity ofvehicles based on airborne image sequences from an infrared camera.At the resolution of approximately one meter vehicles appear as elongatedspots. In urban areas many additional other objects have the sameproperty. To discriminate vehicles from these objects we fuse informationfrom IR-images and vector-maps. Also, evidence from overlapping areasof frames or an IR-sequence is accumulated in the scene. We alsoperform grouping of vehicles into rows along the margins of the roads.These tasks can be automatically performed by a production net. Thegeneric productions for grouping become feasible through the useof map knowledge as context     «

In this contribution we describe a method to assess the activity ofvehicles based on airborne image sequences from an infrared camera.At the resolution of approximately one meter vehicles appear as elongatedspots. In urban areas many additional other objects have the sameproperty. To discriminate vehicles from these objects we fuse informationfrom IR-images and vector-maps. Also, evidence from overlapping areasof frames or an IR-sequence is accumulated in the scene. We alsoperform grouping of ve...     »