Automatic generation of 3D models from real multisensor data

The demand for the prevention of collateral damages in military operations requires methods of robust automatic identification of targets like vehicles especially during target approach. This requires the development of sophisticated techniques for automatic and semi-automatic interpretation of sensor data. In particular, the pre-analysis of reconnaissance data is important for the human observers as well as for autonomous systems. A concept of generating a precise and up-to-date 3D model of a target will be proposed. The model is created from nearterm laser range and infrared data gathered by reconnaissance in advance. It consists of the visible part of the object surface textured with measured infrared values. The model is evaluated by generating 2D templates for a correlation-based tracking system in an infrared image sequence of a complex scenario.