• Home
  • Search
  • O2SNet: Unsupervised Ship Instance Segmentation from Optical to SAR Images Via Domain Adaptation
  • Cite Icon1
  • https://doi.org/10.1109/igarss46834.2022.9883608Copy DOI Icon

O2SNet: Unsupervised Ship Instance Segmentation from Optical to SAR Images Via Domain Adaptation

  • Jul 17, 2022
  • Yiren Wang +2 more
Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Considering insufficient data and difficulty of labeling in Synthetic Aperture Radar (SAR) images, we propose a method for unsupervised SAR ship instance segmentation based on domain adaptation. Compared with optical images, unsupervised learning in SAR images faces the difficulties of insufficient data and lacking detail features. The proposed method O2SNet contains both sample and model adaptation module. The former builds GeminiGAN to convert optical data into SAR-style training samples. In addition, the latter uses optical target features in source domain to guide the unsupervised learning of the SAR images in target domain through prototype alignment. The O2SNet proposed in this paper constructs a novel domain adaptation framework. Experiments show that it has achieved 25.9 and 25.3 points higher Average Precision (AP) in unsupervised detection and segmentation tasks on SAR dataset of HRSID, which has exceeded state-of-the-art domain adaptation methods.

Similar Papers
  • Conference Article
  • Citations4

Cross-Domain Transfer for Ship Instance Segmentation in SAR Images

  • Jul 11, 2021
  • Chunbo Zhu +4
  • Research Article
  • Citations1

Active Style-Content Dual-Branch Domain Adaptation for Semi-Supervised SAR Object Detection.

  • Jan 01, 2026
  • IEEE transactions on image processing : a publication of the IEEE Signal Processing Society
  • Xi Yang +3
  • Research Article
  • Citations33

Hierarchical Similarity Alignment for Domain Adaptive Ship Detection in SAR Images

  • Jan 01, 2022
  • IEEE Transactions on Geoscience and Remote Sensing
  • Jun Zhang +4
  • Research Article
  • Citations1

SAR Ship Detection Based on Deep Domain Adaptation with Limited Samples

  • Jan 01, 2023
  • Procedia Computer Science
  • Shangyi Liu +5
  • PDF
  • Research Article
  • Citations32

SAR Target Detection Based on Domain Adaptive Faster R-CNN with Small Training Data Size

  • Oct 20, 2021
  • Remote Sensing
  • Yuchen Guo +2
  • PDF
  • Research Article
  • Citations1

MtAD-Net: Multi-Threshold Adaptive Decision Net for Unsupervised Synthetic Aperture Radar Ship Instance Segmentation

  • Feb 09, 2025
  • Remote Sensing
  • Junfan Xue +2
  • Research Article
  • Citations39

A Feature Decomposition-Based Method for Automatic Ship Detection Crossing Different Satellite SAR Images

  • Jan 01, 2022
  • IEEE Transactions on Geoscience and Remote Sensing
  • Siyuan Zhao +4
  • Research Article
  • Citations36

An Automatic Ship Detection Method Adapting to Different Satellites SAR Images With Feature Alignment and Compensation Loss

  • Jan 01, 2022
  • IEEE Transactions on Geoscience and Remote Sensing
  • Siyuan Zhao +3
  • PDF
  • Research Article
  • Citations15

Ship Detection from X-Band SAR Images Using M2Det Deep Learning Model

  • Nov 02, 2020
  • Applied Sciences
  • Seong-Jae Hong +2
  • Research Article
  • Citations4

Denoising of SAR images

  • Nov 30, 2011
  • Università degli Studi di Napoli Federico II
  • Mariana Poderico
  • Research Article
  • Citations12

Unsupervised Domain Adaptation for Ship Classification via Progressive Feature Alignment: From Optical to SAR Images

  • Jan 01, 2024
  • IEEE Transactions on Geoscience and Remote Sensing
  • Yu Shi +4
  • Research Article
  • Citations14

Transformer-Based Synthetic-to-Measured SAR Image Translation via Learning of Representational Features

  • Jan 01, 2023
  • IEEE Transactions on Geoscience and Remote Sensing
  • Geunhyuk Youk +1
  • Research Article
  • Citations9

Gradient Prior Guidance and Image Adaptation Enhancement for Semi-Supervised SAR Ship Instance Segmentation

  • Nov 01, 2024
  • IEEE Sensors Journal
  • Man Chen +5
  • Research Article
  • Citations13

Triplet Markov fields with edge location for fast unsupervised multi-class segmentation of synthetic aperture radar images

  • Oct 01, 2012
  • IET Image Processing
  • L Gan +5
  • Research Article
  • Citations5

High-resolution SAR images segmentation using NSCT denoising and QIGA based parameters selection of PCNN model

  • Aug 11, 2020
  • Multimedia Tools and Applications
  • Tian-Jian Luo
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.