提升水下声呐图像检测模型抗噪能力，适用于自主潜水器
ROSAR: An Adversarial Re-Training Framework for Robust Side-Scan Sonar Object Detection
- 结合知识蒸馏与对抗重训练，增强模型鲁棒性
- 在真实声呐数据上实现最高1.85%的检测精度提升
- 提供三个新公开数据集，支持声呐安全属性研究
本文提出ROSAR框架，旨在提升面向侧扫声呐（SSS）图像的深度学习目标检测模型的鲁棒性，该图像由使用声呐传感器的自主水下航行器生成。在先前知识蒸馏（KD）工作的基础上，本框架将KD与对抗重训练相结合，以应对模型效率与抗声呐噪声干扰的双重挑战。我们构建了三个新的公开可用的SSS数据集，涵盖不同声呐配置和噪声条件。提出了两个全新的SSS安全属性，并据此生成对抗样本用于重训练。通过对比投影梯度下降（PGD）与基于块的对抗攻击方法，ROSAR在特定声呐条件下显著提升了模型鲁棒性和检测精度，鲁棒性最高提升达1.85%。代码已开源：https://github.com/remaro-network/ROSAR-framework。
原文摘要 · Abstract (English)
This paper introduces ROSAR, a novel framework enhancing the robustness of deep learning object detection models tailored for side-scan sonar (SSS) images, generated by autonomous underwater vehicles using sonar sensors. By extending our prior work on knowledge distillation (KD), this framework integrates KD with adversarial retraining to address the dual challenges of model efficiency and robustness against SSS noises. We introduce three novel, publicly available SSS datasets, capturing different sonar setups and noise conditions. We propose and formalize two SSS safety properties and utilize them to generate adversarial datasets for retraining. Through a comparative analysis of projected gradient descent (PGD) and patch-based adversarial attacks, ROSAR demonstrates significant improvements in model robustness and detection accuracy under SSS-specific conditions, enhancing the model's robustness by up to 1.85%. ROSAR is available at https://github.com/remaro-network/ROSAR-framework.
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。