CVJan 30, 2024

Towards Assessing the Synthetic-to-Measured Adversarial Vulnerability of SAR ATR

arXiv:2401.17038v119 citationsh-index: 12Has CodeIsprs Journal of Photogrammetry and Remote Sensing
Originality Incremental advance
AI Analysis

It addresses adversarial risks in a practical scenario for remote sensing and computer vision applications, but it is incremental as it builds on existing transfer attack settings.

This paper tackles the vulnerability of deep neural network-based synthetic aperture radar automatic target recognition to adversarial attacks in a synthetic-to-measured transfer setting, where adversarial perturbations generated from synthetic data are transferred to models trained on measured data, and it proposes the transferability estimation attack (TEA) that outperforms state-of-the-art methods and enhances various attack algorithms.

Recently, there has been increasing concern about the vulnerability of deep neural network (DNN)-based synthetic aperture radar (SAR) automatic target recognition (ATR) to adversarial attacks, where a DNN could be easily deceived by clean input with imperceptible but aggressive perturbations. This paper studies the synthetic-to-measured (S2M) transfer setting, where an attacker generates adversarial perturbation based solely on synthetic data and transfers it against victim models trained with measured data. Compared with the current measured-to-measured (M2M) transfer setting, our approach does not need direct access to the victim model or the measured SAR data. We also propose the transferability estimation attack (TEA) to uncover the adversarial risks in this more challenging and practical scenario. The TEA makes full use of the limited similarity between the synthetic and measured data pairs for blind estimation and optimization of S2M transferability, leading to feasible surrogate model enhancement without mastering the victim model and data. Comprehensive evaluations based on the publicly available synthetic and measured paired labeled experiment (SAMPLE) dataset demonstrate that the TEA outperforms state-of-the-art methods and can significantly enhance various attack algorithms in computer vision and remote sensing applications. Codes and data are available at https://github.com/scenarri/S2M-TEA.

Code Implementations1 repo
Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes