Paper 14029-12
Paper 14029-12
Synthetic nfrared (IR) data training evaluation framework for AI/ML object detectors
28 April 2026 • 4:00 PM - 4:20 PM EDT | National Harbor 10
Abstract
Infrared (IR) detection and tracking technologies are advancing rapidly with the integration of artificial intelligence and machine learning (AI/ML). While these tools enhance capability and automation, they also present challenges in ensuring consistent and transparent system performance across diverse operational conditions. Reliable AI/ML training depends on large, high-quality, well curated, datasets, yet real IR data are often limited. Synthetic data offers a promising solution, but their influence on model performance and trustworthiness remains uncertain. This work compares AI object detectors trained using the empirical ATR Algorithm Development Image Database (ADID) and a synthetic counterpart Leveraging explainable AI (XAI) tools the study evaluates how synthetic data may affect model decisions. Results contribute to a framework for validating AI/ML systems that use synthetic data, promoting greater transparency, reliability, and confidence in future IR sensing applications.
Presenter
Torch Technologies, Inc. (United States)
Josh Walters holds both a Bachelor’s in Optical Engineering and Master’s degree in Electrical Engineering from the University of Alabama in Huntsville. He specializes in infrared modeling, simulation, and sensor development. Mr. Walters has experience using the Common Scene Generator (CSG) to create realistic IR scenes that support hardware-in-the-loop, all-digital testing and sensor performance evaluation. He has also developed and implemented detailed IR sensor models capturing optical performance, detector behavior, and signal processing. His work has supported multiple active protection system programs focused on ground and aircraft survivability, where he modeled sensor and threat interactions to enhance detection and countermeasure performance.