Efficient Algorithms for Glint Reduction and Target Tracking in Shipboard IRST

In Phase I we developed novel adaptive spatial-temporal techniques for glint/clutter rejection and image stabilization as well as simplified target detection and tracking methods. As demonstrated in Phase I, our spatial-temporal methods are particularly efficient for target detection/tracking in difficult scenarios when spatial-only filters completely fail. The Phase II proposal emphasizes the development of novel, efficient multisensor and multispectral fusion techniques that will give a substantial performance improvement by using information from several spectral bands; and the development of multispectral target detection and tracking algorithms for reliable detection of extended targets with unknown shapes/signatures. In addition, we propose to extract contours of the objects using edge-motion segmentation techniques. Contour extraction of the object is an important step toward improved ability to track in the presence of sea glint. Contour capabilities allow for multiple feedback loops that improve tracking stability and facilitate reacquisition when a target is dropped. The GUI software will constitute a powerful tool for the optimization of operational target detection/tracking systems in conditions of interest and will be capable of predicting the performance and reducing to a minimum the time needed to search for an “optimal” system configuration in NAVY programs/systems.