用自编码器去除对抗噪声，恢复目标检测性能
Autoencoder-based Denoising Defense against Adversarial Attacks on Object Detection
- 用单层卷积自编码器过滤对抗性扰动
- 检测精度提升3.7%，mAP@50提高10.8%
- 无需重训练模型，适合实时防御场景
基于深度学习的目标检测模型在自动驾驶和安防监控等实际应用中至关重要，但易受对抗样本攻击。本文提出一种基于自编码器的去噪防御方法，以恢复被对抗扰动破坏的检测性能。在COCO数据集的车辆相关图像上，使用分形噪声进行对抗攻击，使bbox mAP从0.2890降至0.1640，性能下降43.3%。通过应用单层卷积自编码器去除扰动后，bbox mAP回升至0.1700（恢复3.7%），bbox mAP@50从0.2780提升至0.3080（提升10.8%）。结果表明，该方法可在不重训练模型的前提下，为对抗攻击提供部分有效防御。
原文摘要 · Abstract (English)
Deep learning-based object detection models play a critical role in real-world applications such as autonomous driving and security surveillance systems, yet they remain vulnerable to adversarial examples. In this work, we propose an autoencoder-based denoising defense to recover object detection performance degraded by adversarial perturbations. We conduct adversarial attacks using Perlin noise on vehicle-related images from the COCO dataset, apply a single-layer convolutional autoencoder to remove the perturbations, and evaluate detection performance using YOLOv5. Our experiments demonstrate that adversarial attacks reduce bbox mAP from 0.2890 to 0.1640, representing a 43.3% performance degradation. After applying the proposed autoencoder defense, bbox mAP improves to 0.1700 (3.7% recovery) and bbox mAP@50 increases from 0.2780 to 0.3080 (10.8% improvement). These results indicate that autoencoder-based denoising can provide partial defense against adversarial attacks without requiring model retraining.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。