Evaluating the use of physics-based modelling to support studies into reflective hyperspectral sensor target detection performance

A first principle physics-based model CameoSim has been used to model a real world woodland scene. Synthetic hyperspectral images have been generated that simulate a typical reflective hyperspectral sensor on an aerial platform. Standard atmospheric correction and detection algorithms have been applied. The results show that target detection can be influenced by many factors in non-intuitive ways, and whilst physics-based modeling may not capture all physical phenomena, it can be a useful tool in the study of hyperspectral target detection.