Robust Feature Extraction and Sensor Fusion for Land Mine Detection

The objective of this project is to develop robust methods for the automatic detection of landmines. During Phase I, we concentrated on developing ATR for the acoustic system from University of Mississippi. We developed a model-based approach that utilizespole properties of mines. The proposed approach offers another mine feature that has not been explored in other approaches. In Phase II we will focus on the following tasks: (1) continue development of the model-based approach to acoustic featureextraction, (2) develop mixed-effects models for statistical mine detection, (3) implement Boosting procedure for fusion of multiple features, and (4) apply Perceptual Organization method to object detection. These methods will be developed based onparticular platforms. Nevertheless, similar ideas can be applied and tested on other platforms as well. The project team consists of Scientific Systems Company, Inc. (SSCI), Planning Systems Inc. (PSI) and the University of Mississippi. The participationof PSI and the University of Mississippi also ensures a smooth transition of the techniques developed under this project.Immediate benefits arising from the research are in the improvement of mine detection performance and speed of advancement for thefuture mine detection systems. The automatic nature of the detection process also help lower training cost and facilitate wide spread use of the new systems. Commercial applications for the developed technology can be realized on any products that requireautomatic object recognition such as surveillance, land-use surveys, and resource management.