Paper 14031-19
Paper 14031-19
Inexpensive adversarially trained infrared panels for evading automated target recognition
28 April 2026 • 4:40 PM - 5:00 PM EDT | National Harbor 10
Abstract
Modern automated target recognition (ATR) systems can use a diverse set of modalities to identify targets. Of particular importance are infrared emissions for night-time detection. In this paper, we describe a physically implemented infrared camouflage system that uses a limited number of temperature-adjustable panels that can be placed on a vehicle and adversarially trained to evade ATR by changing the infrared emitted by the panels.
Testing with a YOLO ATR with and without the camouflage activated, we consistently degrade detection from 90% to below 52% confidence. This was achievable even when only activating 14 of the 24 available panels. This improves upon previous work by creating a cost-effective solution that is optimized to avoid ATR instead of attempting to modify the infrared pattern of the entire target.
Presenter
Constantin Serban
Peraton Labs (United States)
Constantin Serban is a Senior Research Director at Peraton Labs where he leads a research group performing AI/ML data analytics research with application in various domains including autonomy, cyber-security, and network research. Dr. Serban led the development of AI/ML models designed to increase the autonomy of platforms and processes under different settings; he created systems for assessing vulnerabilities in AI/ML algorithms for multiple modalities, including evasion and data poisoning, stretching end-to-end pipelines (sensors to decision) and full AI/ML life-cycle; his research also addresses methods to increase the robustness of AI/ML methods facing such disruptions. He holds a Ph.D. in Computer Science from Rutgers University.