Machine Learning Algorithms for Infrared Search and Track Applications

Toyon proposes research and development of efficient IRST detection and tracking solutions leveraging modern AI/ML techniques. The resulting automated processing technology will enable early warning of long-range threats. Using efficient AI/ML architectures we will provide significant improvements in computational efficiency while maintaining or improving the accuracy of dim target detection and tracking provided by existing algorithms. We will demonstrate feasibility and evaluate the performance of AI/ML-based IRST in comparison with a state-of-the-art baseline IRST processing chain. A mix of real and synthetic IRST data will be used to train models, and evaluations will be performed using sequestered real data. The resulting technology will enable IRST processing with low cost, size, weight, and power (C-SWaP).