Unsupervised Domain Adaptation for SAR Ship Detection Based on Multitask Decoupling

Remote sensing ship detection is crucial for maritime surveillance, especially with the rise of synthetic aperture radar (SAR) images, known for their all-weather capabilities. Yet, annotating SAR images is resource-intensive and requires specialized expertise. Facing the challenge, we propose an un...

Full description

Saved in:
Bibliographic Details
Main Authors: Yirong Yang, Xi Yang, Dong Yang
Format: Article
Language:English
Published: IEEE 2025-01-01
Series:IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Subjects:
Online Access:https://ieeexplore.ieee.org/document/10993287/
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849717616717856768
author Yirong Yang
Xi Yang
Dong Yang
author_facet Yirong Yang
Xi Yang
Dong Yang
author_sort Yirong Yang
collection DOAJ
description Remote sensing ship detection is crucial for maritime surveillance, especially with the rise of synthetic aperture radar (SAR) images, known for their all-weather capabilities. Yet, annotating SAR images is resource-intensive and requires specialized expertise. Facing the challenge, we propose an unsupervised domain adaptation method for SAR ship detection based on multitask decoupling, which leverages data annotated with optical images instead of SAR annotations. In domain adaptation object detection, conflict arises between transferability required for adaptation and discriminability required for detection. Similarly, within object detection, conflict in feature preferences between classification and regression tasks occurs. We address these conflicts by decoupling the adaptation module from the detector module and decoupling the adaptation module into category, and bounding box adaptation modules. Our detector module integrates a feature enhancement network (FEN) to improve the classification and localization ability under the challenge of dense small targets with low visual contrast in SAR images. The adaptation module employs adversarial adaptation and cross-domain category alignment (CCA) to pull up the feature representations of different domains of ships for domain alignment. In addition, the cross-domain image synthesis (CDIS) module generates transitional images covering SAR domain style and optical domain content to bridge the huge gap between the domains. Through the separation and coordination of different modules, our multitask decoupling framework strategically balances feature transferability and discriminability. Extensive experiments on classic remote sensing datasets validate the effectiveness of our method, showing better performance than state-of-the-art methodologies.
format Article
id doaj-art-1f84254a53bd48faba0d1a3817948657
institution DOAJ
issn 1939-1404
2151-1535
language English
publishDate 2025-01-01
publisher IEEE
record_format Article
series IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
spelling doaj-art-1f84254a53bd48faba0d1a38179486572025-08-20T03:12:36ZengIEEEIEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing1939-14042151-15352025-01-0118126841269610.1109/JSTARS.2025.356821210993287Unsupervised Domain Adaptation for SAR Ship Detection Based on Multitask DecouplingYirong Yang0Xi Yang1https://orcid.org/0000-0002-5791-3674Dong Yang2State Key Laboratory of Integrated Services Networks, School of Telecommunications Engineering, Xidian University, Xi'an, ChinaState Key Laboratory of Integrated Services Networks, School of Telecommunications Engineering, Xidian University, Xi'an, ChinaXi'an Institute of Space Radio Technology, Xi'an, ChinaRemote sensing ship detection is crucial for maritime surveillance, especially with the rise of synthetic aperture radar (SAR) images, known for their all-weather capabilities. Yet, annotating SAR images is resource-intensive and requires specialized expertise. Facing the challenge, we propose an unsupervised domain adaptation method for SAR ship detection based on multitask decoupling, which leverages data annotated with optical images instead of SAR annotations. In domain adaptation object detection, conflict arises between transferability required for adaptation and discriminability required for detection. Similarly, within object detection, conflict in feature preferences between classification and regression tasks occurs. We address these conflicts by decoupling the adaptation module from the detector module and decoupling the adaptation module into category, and bounding box adaptation modules. Our detector module integrates a feature enhancement network (FEN) to improve the classification and localization ability under the challenge of dense small targets with low visual contrast in SAR images. The adaptation module employs adversarial adaptation and cross-domain category alignment (CCA) to pull up the feature representations of different domains of ships for domain alignment. In addition, the cross-domain image synthesis (CDIS) module generates transitional images covering SAR domain style and optical domain content to bridge the huge gap between the domains. Through the separation and coordination of different modules, our multitask decoupling framework strategically balances feature transferability and discriminability. Extensive experiments on classic remote sensing datasets validate the effectiveness of our method, showing better performance than state-of-the-art methodologies.https://ieeexplore.ieee.org/document/10993287/Domain adaptationremote sensing imagessynthetic aperture radar (SAR) imagesship detection
spellingShingle Yirong Yang
Xi Yang
Dong Yang
Unsupervised Domain Adaptation for SAR Ship Detection Based on Multitask Decoupling
IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Domain adaptation
remote sensing images
synthetic aperture radar (SAR) images
ship detection
title Unsupervised Domain Adaptation for SAR Ship Detection Based on Multitask Decoupling
title_full Unsupervised Domain Adaptation for SAR Ship Detection Based on Multitask Decoupling
title_fullStr Unsupervised Domain Adaptation for SAR Ship Detection Based on Multitask Decoupling
title_full_unstemmed Unsupervised Domain Adaptation for SAR Ship Detection Based on Multitask Decoupling
title_short Unsupervised Domain Adaptation for SAR Ship Detection Based on Multitask Decoupling
title_sort unsupervised domain adaptation for sar ship detection based on multitask decoupling
topic Domain adaptation
remote sensing images
synthetic aperture radar (SAR) images
ship detection
url https://ieeexplore.ieee.org/document/10993287/
work_keys_str_mv AT yirongyang unsuperviseddomainadaptationforsarshipdetectionbasedonmultitaskdecoupling
AT xiyang unsuperviseddomainadaptationforsarshipdetectionbasedonmultitaskdecoupling
AT dongyang unsuperviseddomainadaptationforsarshipdetectionbasedonmultitaskdecoupling