R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive Aggregation

Synthetic Aperture Radar (SAR) is extensively utilized in ship detection due to its robust performance under various weather conditions and its capability to operate effectively both during the day and at night. However, ships in SAR images exhibit various characteristics including complex land scat...

Full description

Saved in:
Bibliographic Details
Main Authors: Xiaoting Li, Wei Duan, Xikai Fu, Xiaolei Lv
Format: Article
Language:English
Published: MDPI AG 2025-02-01
Series:Remote Sensing
Subjects:
Online Access:https://www.mdpi.com/2072-4292/17/3/551
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850199776138625024
author Xiaoting Li
Wei Duan
Xikai Fu
Xiaolei Lv
author_facet Xiaoting Li
Wei Duan
Xikai Fu
Xiaolei Lv
author_sort Xiaoting Li
collection DOAJ
description Synthetic Aperture Radar (SAR) is extensively utilized in ship detection due to its robust performance under various weather conditions and its capability to operate effectively both during the day and at night. However, ships in SAR images exhibit various characteristics including complex land scattering interference, variable scales, and dense spatial arrangements. Existing algorithms are insufficient in effectively addressing these challenges. To enhance detection accuracy, this paper proposes the Rotated model with Spatial Aggregation and a Balanced-Shifted Mechanism (R-SABMNet) built upon YOLOv8. First, we introduce the Spatial-Guided Adaptive Feature Aggregation (SG-AFA) module, which enhances sensitivity to ship features while suppressing land scattering interference. Subsequently, we propose the Balanced Shifted Multi-Scale Fusion (BSMF) module, which effectively enhances local detail information and improves adaptability to multi-scale targets. Finally, we introduce the Gaussian Wasserstein Distance Loss (GWD), which effectively addresses localization errors arising from angle and scale inconsistencies in dense scenes. Our R-SABMNet outperforms other deep learning-based methods on the SSDD+ and HRSID datasets. Specifically, our method achieves a detection accuracy of 96.32%, a recall of 93.13%, and an average level of accuracy of 95.28% on the SSDD+ dataset.
format Article
id doaj-art-bbdb2a0fec15401bbcaa492bc6a6048f
institution OA Journals
issn 2072-4292
language English
publishDate 2025-02-01
publisher MDPI AG
record_format Article
series Remote Sensing
spelling doaj-art-bbdb2a0fec15401bbcaa492bc6a6048f2025-08-20T02:12:33ZengMDPI AGRemote Sensing2072-42922025-02-0117355110.3390/rs17030551R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive AggregationXiaoting Li0Wei Duan1Xikai Fu2Xiaolei Lv3School of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, ChinaInstitute of Software, Chinese Academy of Sciences, Beijing 100190, ChinaKey Laboratory of Technology in Geo-Spatial Information Processing and Application System, Chinese Academy of Sciences, Beijing 100190, ChinaSchool of Electronic, Electrical and Communication Engineering, University of Chinese Academy of Sciences, Beijing 100049, ChinaSynthetic Aperture Radar (SAR) is extensively utilized in ship detection due to its robust performance under various weather conditions and its capability to operate effectively both during the day and at night. However, ships in SAR images exhibit various characteristics including complex land scattering interference, variable scales, and dense spatial arrangements. Existing algorithms are insufficient in effectively addressing these challenges. To enhance detection accuracy, this paper proposes the Rotated model with Spatial Aggregation and a Balanced-Shifted Mechanism (R-SABMNet) built upon YOLOv8. First, we introduce the Spatial-Guided Adaptive Feature Aggregation (SG-AFA) module, which enhances sensitivity to ship features while suppressing land scattering interference. Subsequently, we propose the Balanced Shifted Multi-Scale Fusion (BSMF) module, which effectively enhances local detail information and improves adaptability to multi-scale targets. Finally, we introduce the Gaussian Wasserstein Distance Loss (GWD), which effectively addresses localization errors arising from angle and scale inconsistencies in dense scenes. Our R-SABMNet outperforms other deep learning-based methods on the SSDD+ and HRSID datasets. Specifically, our method achieves a detection accuracy of 96.32%, a recall of 93.13%, and an average level of accuracy of 95.28% on the SSDD+ dataset.https://www.mdpi.com/2072-4292/17/3/551ship detectionland scattering interferencesynthetic aperture radar (SAR)multi-head self-attentiontwo-dimensional Gaussian distribution
spellingShingle Xiaoting Li
Wei Duan
Xikai Fu
Xiaolei Lv
R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive Aggregation
Remote Sensing
ship detection
land scattering interference
synthetic aperture radar (SAR)
multi-head self-attention
two-dimensional Gaussian distribution
title R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive Aggregation
title_full R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive Aggregation
title_fullStr R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive Aggregation
title_full_unstemmed R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive Aggregation
title_short R-SABMNet: A YOLOv8-Based Model for Oriented SAR Ship Detection with Spatial Adaptive Aggregation
title_sort r sabmnet a yolov8 based model for oriented sar ship detection with spatial adaptive aggregation
topic ship detection
land scattering interference
synthetic aperture radar (SAR)
multi-head self-attention
two-dimensional Gaussian distribution
url https://www.mdpi.com/2072-4292/17/3/551
work_keys_str_mv AT xiaotingli rsabmnetayolov8basedmodelfororientedsarshipdetectionwithspatialadaptiveaggregation
AT weiduan rsabmnetayolov8basedmodelfororientedsarshipdetectionwithspatialadaptiveaggregation
AT xikaifu rsabmnetayolov8basedmodelfororientedsarshipdetectionwithspatialadaptiveaggregation
AT xiaoleilv rsabmnetayolov8basedmodelfororientedsarshipdetectionwithspatialadaptiveaggregation