ROSAR: An Adversarial Re-Training Framework for Robust Side-Scan Sonar Object Detection
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866914971654291456 |
|---|---|
| author | Aubard, Martin Antal, László Madureira, Ana Teixeira, Luis F. Ábrahám, Erika |
| author_facet | Aubard, Martin Antal, László Madureira, Ana Teixeira, Luis F. Ábrahám, Erika |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_10554 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ROSAR: An Adversarial Re-Training Framework for Robust Side-Scan Sonar Object Detection Aubard, Martin Antal, László Madureira, Ana Teixeira, Luis F. Ábrahám, Erika Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Robotics 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. |
| title | ROSAR: An Adversarial Re-Training Framework for Robust Side-Scan Sonar Object Detection |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning Robotics |
| url | https://arxiv.org/abs/2410.10554 |