Unsupervised SAR Fine-Grained Ship Classification via Spherical Metric Refinement With Deep Subdomain Adaptation

Unsupervised domain adaptation (UDA) is a promising method for addressing the problem of SAR fine-grained ship classification in target domain with no labeled data available by leveraging a large number of labeled samples from source domains. This article proposes a novel framework, spherical metric...

Full description

Saved in:
Bibliographic Details
Main Authors: Zhichao Han, Haitao Lang
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/11045289/
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Unsupervised domain adaptation (UDA) is a promising method for addressing the problem of SAR fine-grained ship classification in target domain with no labeled data available by leveraging a large number of labeled samples from source domains. This article proposes a novel framework, spherical metric refinement with deep subdomain adaptation, to address two crucial issues that are rarely recognized by existing UDA approaches, namely prioritizing <italic>adaptation</italic> over fine-grained <italic>classification</italic> and hindering cross-domain alignment and discrimination due to Euclidean feature norms. The proposed solution transforms features into spherical space to eliminate norm bias and introduces joint optimization of <italic>classification</italic> and <italic>adaptation</italic>, balancing discriminative feature learning and domain invariance. Experiments on GF-SAR and HR-SAR datasets demonstrate state-of-the-art performance, achieving 95.33&#x0025; and 89.33&#x0025; classification accuracy, respectively, outperforming the existing methods by 5.33&#x2013;6.00&#x0025;. Our GF-SAR and HR-SAR datasets have been released on GitHub.
ISSN:1939-1404
2151-1535