Zhang et al. (2026) Estimating wheat lodging fraction from Sentinel-1/2 imagery using UAV-derived reference labels

Identification

• Journal: Computers and Electronics in Agriculture
• Year: 2026
• Date: 2026-09-11
• Authors: Baoyuan Zhang, Tianxiang Zhang, Xingyu Liu, Chao Song, Xiaoyuan Bao, Qian Sun, Yuchun Pan, Xia Yao, Meiyan Shu, Xiaohe Gu
• DOI: 10.1016/j.compag.2026.112410

Research Groups

• College of Smart Agriculture, Nanjing Agricultural University, China
• Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, China
• Ping An Technology (Shenzhen) Co., Ltd., Shenzhen, China
• State Key Laboratory of Aridland Crop Science, Gansu Agricultural University, Lanzhou, China
• Jiangsu Co-Innovation Center for Modern Production Technology of Grain Crops, Yangzhou University, Yangzhou, China
• Collaborative Innovation Center for Modern Crop Production co-sponsored by Province and Ministry, Nanjing, China
• College of Information and Management Science, Henan Agricultural University, Zhengzhou, China

Short Summary

This study developed a framework to estimate wheat lodging fraction (WLF) from Sentinel-1/2 imagery using UAV-derived reference labels. The proposed approach achieved high accuracy in estimating WLF with an R2 value of 0.825 and NRMSE of 12.60%.

Objective

• Estimate wheat lodging fraction (WLF) from Sentinel-1/2 imagery using UAV-derived reference labels.

Study Configuration

• Spatial Scale: 10 × 10 m grids, covering winter wheat-growing areas in Xinxiang City, China.
• Temporal Scale: Multi-temporal remote sensing data, including pre- and post-event optical and SAR data.

Methodology and Data

• Models used: Feature Tokenizer Transformer (FT-Transformer) with combined Sentinel-1/2 pre–post changes.
• Data sources: UAV-derived reference labels, Sentinel-1/2 imagery, and ground truth data from field surveys.

Main Results

• The proposed approach achieved high accuracy in estimating WLF, with an R2 value of 0.825 and NRMSE of 12.60% on the grouped cross-validation subset.
• On the spatially independent test subset, the model achieved R2 = 0.770, RMSE = 0.117, and NRMSE = 12.11%.

Contributions

• The study provides a reproducible framework for extending spatially distributed UAV observations to event-scale satellite WLF assessment.
• The proposed approach enables rapid and accurate estimation of WLF from Sentinel-1/2 imagery, which can be used for decision-making in agriculture.

Funding

• This research was supported by the National Natural Science Foundation of China (Grant No. 52177044) and the Jiangsu Co-Innovation Center for Modern Production Technology of Grain Crops (Grant No. JS2019003).

Citation

@article{Zhang2026Estimating,
  author = {Zhang, Baoyuan and Zhang, Tianxiang and Liu, Xingyu and Song, Chao and Bao, Xiaoyuan and Sun, Qian and Pan, Yuchun and Yao, Xia and Shu, Meiyan and Gu, Xiaohe},
  title = {Estimating wheat lodging fraction from Sentinel-1/2 imagery using UAV-derived reference labels},
  journal = {Computers and Electronics in Agriculture},
  year = {2026},
  doi = {10.1016/j.compag.2026.112410},
  url = {https://doi.org/10.1016/j.compag.2026.112410}
}