ML Security PapersKnowledge-Guided Adversarial Training for Infrared Object Detection via Thermal Radiation Modeling
Shiji Zhao 1, Shukun Xiong 1, Maoxun Yuan 1, Yao Huang 1, Ranjie Duan 2, Qing Guo 3, Jiansheng Chen 4, Haibin Duan 1, Xingxing Wei 1
Published on arXiv
2603.25170
Input Manipulation Attack
OWASP ML Top 10 — ML01
Key Finding
KGAT enhances both clean accuracy and robustness against adversarial attacks and common corruptions on three infrared datasets across six mainstream infrared object detection models
KGAT
Novel technique introduced
In complex environments, infrared object detection exhibits broad applicability and stability across diverse scenarios. However, infrared object detection is vulnerable to both common corruptions and adversarial examples, leading to potential security risks. To improve the robustness of infrared object detection, current methods mostly adopt a data-driven ideology, which only superficially drives the network to fit the training data without specifically considering the unique characteristics of infrared images, resulting in limited robustness. In this paper, we revisit infrared physical knowledge and find that relative thermal radiation relations between different classes can be regarded as a reliable knowledge source under the complex scenarios of adversarial examples and common corruptions. Thus, we theoretically model thermal radiation relations based on the rank order of gray values for different classes, and further quantify the stability of various inter-class thermal radiation relations. Based on the above theoretical framework, we propose Knowledge-Guided Adversarial Training (KGAT) for infrared object detection, in which infrared physical knowledge is embedded into the adversarial training process, and the predicted results are optimized to be consistent with the actual physical laws. Extensive experiments on three infrared datasets and six mainstream infrared object detection models demonstrate that KGAT effectively enhances both clean accuracy and robustness against adversarial attacks and common corruptions.
Key Contributions
- Theoretical framework modeling thermal radiation relations in infrared images based on rank order of gray values across object classes
- Knowledge-Guided Adversarial Training (KGAT) that embeds infrared physical knowledge into adversarial training to enforce consistency with thermal radiation laws
- Demonstrated effectiveness across three infrared datasets and six object detection models for both clean accuracy and robustness against adversarial attacks and common corruptions
🛡️ Threat Analysis
Paper addresses adversarial robustness of infrared object detection models against adversarial attacks at inference time, and proposes adversarial training as a defense mechanism.