Saved in:
Bibliographic Details
Main Authors: Li, Yuyang, Han, Jiashu, Lai, Yinyi, Kang, Wenbin, Liu, Zenghui
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.07388
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910200755126272
author Li, Yuyang
Han, Jiashu
Lai, Yinyi
Kang, Wenbin
Liu, Zenghui
author_facet Li, Yuyang
Han, Jiashu
Lai, Yinyi
Kang, Wenbin
Liu, Zenghui
contents Marine debris detection for ocean robot is crucial for ecological protection, yet performance is often degraded by low-quality images with blur, complex backgrounds, and small targets. To address these challenges, we propose YOLO-MD, an enhanced YOLO-based detection framework. A Dual-Branch Convolutional Enhanced Self-Attention (DB-CASA) module is designed to strengthen spatial-channel interactions, improving feature representation in degraded images. Additionally, a lightweight shift-based operation is introduced to enhance fine-grained feature extraction for objects of varying scales while maintaining parameter efficiency. We further propose SFG-Loss to mitigate class imbalance and optimization instability via dynamic sample reweighting. Experiments on the UODM dataset demonstrate that YOLO-MD achieves 0.875 precision, 0.822 F1-score, and 0.849 mAP50, outperforming the latest state-of-the-art methods. The effectiveness of this method has also been verified through real-world robotic edge deployment experiments.
format Preprint
id arxiv_https___arxiv_org_abs_2605_07388
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Marine Debris Detection Framework for Ocean Robots via Self-Attention Enhancement and Feature Interaction Optimization
Li, Yuyang
Han, Jiashu
Lai, Yinyi
Kang, Wenbin
Liu, Zenghui
Computer Vision and Pattern Recognition
Marine debris detection for ocean robot is crucial for ecological protection, yet performance is often degraded by low-quality images with blur, complex backgrounds, and small targets. To address these challenges, we propose YOLO-MD, an enhanced YOLO-based detection framework. A Dual-Branch Convolutional Enhanced Self-Attention (DB-CASA) module is designed to strengthen spatial-channel interactions, improving feature representation in degraded images. Additionally, a lightweight shift-based operation is introduced to enhance fine-grained feature extraction for objects of varying scales while maintaining parameter efficiency. We further propose SFG-Loss to mitigate class imbalance and optimization instability via dynamic sample reweighting. Experiments on the UODM dataset demonstrate that YOLO-MD achieves 0.875 precision, 0.822 F1-score, and 0.849 mAP50, outperforming the latest state-of-the-art methods. The effectiveness of this method has also been verified through real-world robotic edge deployment experiments.
title A Marine Debris Detection Framework for Ocean Robots via Self-Attention Enhancement and Feature Interaction Optimization
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.07388