Computer vision and sensor fusion for detecting buried objects

Given multiple images of the Earth's surface from dual-band infrared sensors, a system that fuses information from the sensors to reduce the effects of clutter and improve the ability to detect buried or surface target sites is presented. Supervised learning pattern classifiers (including neural networks) are used. Results of experiments to detect buried land mines from real data are given, and the usefulness of fusing information from multiple sensor types is evaluated. The novelty of the work lies mostly in the combination of the algorithms and their application to the very important and currently unsolved problem of detecting buried land mines from an airborne standoff platform. >