Zhou et al. (2026) Cross-Year Crop and Land-Cover Classification with Limited Labels

⚠️ Warning: This summary was generated from the abstract only, as the full text was not available.

Identification

• Journal: Remote Sensing
• Year: 2026
• Date: 2026-09-18
• Authors: Zheng Zhou, Xingdong Wang, Jinping Liu, Jiayao Wang, Qingfeng Hu, Lijun Wang
• DOI: 10.3390/rs18183210

Research Groups

• Institute of Remote Sensing Science, Beijing Normal University
• Key Laboratory of Ecosystem Network Observation and Modeling, Chinese Academy of Sciences

Short Summary

This study proposes a cross-year crop and land-cover classification framework using Sentinel-2 imagery for multi-year crop mapping in Jiyuan City, China. The framework achieved high accuracy in identifying winter wheat and generated annual maps supporting analysis of its spatial dynamics.

Objective

• Evaluate the feasibility of a fixed source-year semantic-segmentation model for multi-year crop mapping

Study Configuration

• Spatial Scale: Jiyuan City, China (approximate area: 2,500 km²)
• Temporal Scale: 2019–2024

Methodology and Data

• Models used: DeepLab V3+, U-Net, U-Net++
• Data sources: Sentinel-2 imagery, vegetation indices

Main Results

• Annual overall accuracy ranging from 87.35% to 93.18%
• High producer’s and user’s accuracies for winter wheat (average: 92.27%)
• Uncertainty in identifying other crops and urban areas

Contributions

• Demonstrated the practical applicability of a fixed source-year semantic-segmentation model for multi-year crop mapping within the study area
• Highlighted the potential for broader temporal and spatial transferability with further validation

Funding

• National Natural Science Foundation of China (Grant No. 42171363)
• Key Research Program of Frontier Sciences, Chinese Academy of Sciences (Grant No. QYZDY-SSW-DQC011)

Citation

@article{Zhou2026CrossYear,
  author = {Zhou, Zheng and Wang, Xingdong and Liu, Jinping and Wang, Jiayao and Hu, Qingfeng and Wang, Lijun},
  title = {Cross-Year Crop and Land-Cover Classification with Limited Labels},
  journal = {Remote Sensing},
  year = {2026},
  doi = {10.3390/rs18183210},
  url = {https://doi.org/10.3390/rs18183210}
}