Robust Algorithms Enhance Target Detection and Aim Point Selection

The proposed Pulse Coupled Neural Network (PCNN) algorithm presents a robust method for quickly analyzing generalized imagery while requiring very few control inputs. Processor bandwidth is reduced by locating and eliminating pixels with no information content before the pixel field is processed. In some cases, this can reduce the downstream bandwidth requirements by as much as two orders of magnitude. The PCNN algorithm has the potential to detect multiple targets within a scene of low contrast data and improve the operation of existing seekers in all imaging applications, while still providing the most cost-effective means of detecting low-intensity targets. Phase I will determine the merits of using the PCNN to remove data from downstream processing for THAAD. Data and expert discussion provided by Lockheed Martin Space Systems Company will ensure the early insertion of this technology for the main line THAAD program and advanced seekers. During Phase II, the PCNN algorithm will be implemented in software and firmware in a massively parallel processor designed specifically for THAAD and the next generation seekers. Polaris Sensor Technologies, Inc. will leverage prior research and development projects to lower technological risk for this MDA effort.