• Home
  • Search
  • A Class-Aware Unsupervised Domain Adaptation Framework for Cross-Continental Crop Classification with Sentinel-2 Time Series
  • Cite Icon1
  • https://doi.org/10.3390/rs17223762Copy DOI Icon

A Class-Aware Unsupervised Domain Adaptation Framework for Cross-Continental Crop Classification with Sentinel-2 Time Series

Show More
  • Abstract
  • Literature Map
  • References
  • Citations
  • Similar Papers
Abstract

Accurate and large-scale crop mapping is crucial for global food security, yet its performance is often hindered by domain shift when models trained in one region are applied to another. This is particularly challenging in cross-continental scenarios where variations in climate, soil, and farming systems are significant. To address this, we propose PLCM (PSAE-LTAE + Class-aware MMD), an unsupervised domain adaptation (UDA) framework for crop classification using Sentinel-2 satellite image time series. The framework features two key innovations: (1) a Pixel-Set Attention Encoder (PSAE), which intelligently aggregates spatial features within parcels by assigning weights to individual pixels, enhancing robustness against noise and intra-parcel heterogeneity; and (2) a class-aware Maximum Mean Discrepancy (MMD) loss function that performs fine-grained feature alignment within each crop category, effectively mitigating negative transfer caused by domain shift while preserving class-discriminative information. We validated our framework on a challenging cross-continental, cross-year task, transferring a model trained on data from the source domain in the United States (2022) to an unlabeled target domain in Wensu County, Xinjiang, China (2024). The results demonstrate the robust performance of PLCM. While achieving a competitive overall Macro F1-score of 96.56%, comparable to other state-of-the-art UDA methods, its primary advantage is revealed in a granular per-class analysis. This analysis shows that PLCM provides a more balanced performance by particularly excelling at identifying difficult-to-adapt categories (e.g., Cotton), demonstrating practical robustness. Ablation studies further confirmed that both the PSAE module and the class-aware MMD strategy were critical to this performance gain. Our study shows that the PLCM framework can effectively learn domain-invariant and class-discriminative features, offering an effective and robust solution for high-accuracy, large-scale crop mapping across diverse geographical regions.

Similar Papers
  • Research Article
  • Citations6

Semantically preserving adversarial unsupervised domain adaptation network for improving disease recognition from chest x-rays.

  • Jul 01, 2023
  • Computerized Medical Imaging and Graphics
  • Xiaoli Qin +2
  • Abstract

Unsupervised Domain Adaptation of Auto-Segmentation on Multi-Source MRIs

  • Sep 29, 2023
  • International Journal of Radiation Oncology*Biology*Physics
  • W Zhang +4
  • Research Article
  • Citations4

Domain-Invariant Feature Learning for Domain Adaptation

  • Mar 03, 2023
  • International Journal of Pattern Recognition and Artificial Intelligence
  • Ching-Ting Tu +4
  • Research Article
  • Citations27

Dual domain distribution disruption with semantics preservation: Unsupervised domain adaptation for medical image segmentation

  • Jul 14, 2024
  • Medical Image Analysis
  • Boyun Zheng +8
  • Research Article
  • Citations6

Cross-Domain Urban Land Use Classification via Scenewise Unsupervised Multisource Domain Adaptation With Transformer

  • Jan 01, 2024
  • IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
  • Mengmeng Li +3
  • PDF
  • Research Article
  • Citations58

The Eyes of the Gods: A Survey of Unsupervised Domain Adaptation Methods Based on Remote Sensing Data

  • Sep 03, 2022
  • Remote Sensing
  • Mengqiu Xu +4
  • Research Article
  • Citations113

Weighted and Class-Specific Maximum Mean Discrepancy for Unsupervised Domain Adaptation

  • Nov 22, 2019
  • IEEE Transactions on Multimedia
  • Hongliang Yan +5
  • Research Article
  • Citations1

Towards Pose Invariant Bionic Limb Control: A comparative study of two unsupervised domain adaptation methods

  • Jan 01, 2025
  • IEEE Access
  • Alexander Hannius +2
  • Research Article
  • Citations3

Unsupervised global-local domain adaptation with self-training for remote sensing image semantic segmentation

  • Jan 20, 2025
  • International Journal of Remote Sensing
  • Junbo Zhang +3
  • Book Chapter

Iterative Discriminative Domain Adaptation

  • Jan 01, 2019
  • Xiaofu Wu +3
  • Research Article
  • Citations36

Unsupervised Domain Adaptation by Matching Distributions Based on the Maximum Mean Discrepancy via Unilateral Transformations

  • Jul 17, 2019
  • Proceedings of the AAAI Conference on Artificial Intelligence
  • Atsutoshi Kumagai +1
  • Research Article
  • Citations41

Structure-Driven Unsupervised Domain Adaptation for Cross-Modality Cardiac Segmentation

  • Dec 01, 2021
  • IEEE Transactions on Medical Imaging
  • Zhiming Cui +8
  • Research Article
  • Citations47

Adapt Everywhere: Unsupervised Adaptation of Point-Clouds and Entropy Minimization for Multi-Modal Cardiac Image Segmentation.

  • Jul 01, 2021
  • IEEE transactions on medical imaging
  • Sulaiman Vesal +4
  • Book Chapter

Unsupervised Domain Adaptation Using Temporal Association for Segmentation and Its Application to C. elegans Time-Lapse Images

  • Jan 01, 2022
  • Hiroaki Nozaki +1
  • Research Article
  • Citations14

Unsupervised deep domain adaptation algorithm for video based human activity recognition via recurrent neural networks

  • Jul 07, 2024
  • Engineering Applications of Artificial Intelligence
  • Abdulaziz Zam +2
Cactus Communications logo

Copyright 2026 Cactus Communications. All rights reserved.