Methodologies for a Partial Frame Correlation of Multiple Sensors

In order to provide the highest level Quality of Service (QoS) for both tracking and discrimination functions it is imperative that information from all sensors be associated correctly. Due to potential bandwidth, latency and timeline impacts, it may be necessary to limit downlinked Airborne Infrared (ABIR) data to portions of the focal plane around exceedances of interest. As a result, we propose to exploit these partial frame chips to aid in the sensor to sensor correlation problem. The author proposes a novel set of algorithms that work in two phases. The first phase will provide robust sensor to sensor correlation in the absence of partial frames using a multiple hypothesis algorithm with the ability to flag cases where solutions are ambiguous. In the cases where ambiguous solutions exist, phase two will exploit the partial frame chips in order to provide a unique solution. The algorithm suite will draw from previously developed image processing algorithms to provide the best possible super-resolution of the partial frames and industry leading correlation algorithms specifically designed for EO/IR assets. Through the use of multi-hypothesis algorithms exercised on partial frames, it will be possible to overcome the ambiguities associated with the sensor to sensor correlation problem.