Chatman Phase 2 Sequential

Reducing the False Alarm Rate (FAR) of Automated Target Recognition (ATR) algorithms for Synthetic Aperture Radar (SAR) imagery is crucial for Intelligence, Surveillance, Reconnaissance (ISR) and precision target engagement missions. While modern Deep Learning (DL) ATR networks have demonstrated advanced predictive capabilities and generalization for SAR imagery, they lack spatial awareness, resulting in a higher FAR. ATR networks are not functionally equipped to learn semantic separability of scene geometries external to the target ontology, hindering their ability to distill contextual relevancy between objects to both increase precision and reduce FAR. To address these network deficiencies, the Stratagem team proposed a solution to the Phase I NGA OSD221-001 SBIR called: Context-aware Hierarchical graph network for improved ATR performance (CHATMAN), a Scene Geometry Aided (SGA) ATR framework which distills relations between external scene geometries and detected targets to reduce false alarms.Our Phase II solution builds upon Phase I by integrating the independent models developed in Phase I into a single, end-to-end trainable framework and enhancing detection capabilities through multimodal data fusion.In the Phase II extension, the Stratagem team will further develop the solution to achieve a more operational state and continue advancing detection capabilities. Our specific goals include: 1. Replacing the ATR component with an encoder network to establish an ATR-agnostic framework; 2. Enhancing detection capabilities through multimodal data fusion with target-specific data sources; 3. Deploying our solution to an operational, government-hosted environment.