Robust Maritime Target Recognition

The proliferation of group 3-5 unmanned aerial systems (UAS) provides an excellent opportunity for robust, passive, maritime target recognition and classification.Ā These platforms can be outfitted with (1) high-quality camera systems, (2) high-performance compute resources, and (3) encrypted links for relaying tactical information to other nodes. RDRTec has already demonstrated the ability to perform maritime target classification from ISAR imagery at very long ranges which separates combatants from commercial vessels wand is able to distinguish combatants at levels of detail below fine naval class.Ā Our toolset and reference database divides ships of the world into over 1500 types.Ā This approach requires no training and performance has been demonstrated on datasets from new sensors with no parameter tuning. RDRTecÆs Maritime Classification Aid (MCA) sorts image frames into categories. MCAÆs profile feature extraction will work with almost no modifications for EO/IR imagery taken from angles +/- 65 deg from broadside and grazing angles up to 15 deg.Ā For these angles, feature extraction identifyies superstructure location and width at various heights.Ā Our database is fully populated for superstructure features for all 1500+ templates.Ā For frames that are not profile views which use RCS-based feature extraction, direct measurement of superstructure locations and sizes are not always directly measurable. MCA has demonstrated that we can approximate the expected extracted features from these viewing angles right from a single profile view reference image and perform classification without any 3D information. RDRTecÆs MCA has already been integrated with mission systems such as the USNÆs Minotaur system, and sensor resource managers such as Lambda ScienceÆs SEMS.Ā It has been hosted on flight ready tactical processors.Ā Part of MCAÆs design requirements has always been the ability to run in real-time on modest processor resources.Ā Currently we have a delivered configuration that can run in a single 3U Open-VPX slot while performing not only classification but also performs ISAR image formation starting from raw IQ data. RDRTec is currently advancing a hybrid physics-based / ML addition to MCA for exploiting vibration and rotating aperture signatures and will take a similar approach in augmenting MCA for use with EO/IR sensors across wide viewing angles. Keys to RDRTecÆs approach include (1) proven performance on ISAR imagery, (2) existing database of ships of the world, (3) prior integrations with mission systems and sensor resource managers, (4) no training required, (5) very conservative SWAP, (6) successful transitions as a supplier of third-party modes/apps to DoD primes and DoD labs.