Synergetic Integration Of Thermal And Visual Images For Computer Vision

This paper presents a new computer vision technique based on multisensor fusion for interpreting images of outdoor scenes. The technique is based on the synergetic integration of information derived from thermal and visual sensors. Information integration is implemented at different levels of abstraction in the interpretation hierarchy i.e., at the pixel and at the symbolic levels. The approach establishes features based on surface heat flux estimates that describe intrinsic thermal behavior of imaged objects. Heuristic rules are employed in a decision tree classifier to categorize imaged objects as being either vegetation, building, pavement or a vehicle.