Wang et al. (2026) Estimation of cotton plant moisture content using UAV multimodal data and machine learning

Identification

• Journal: Agricultural Water Management
• Year: 2026
• Date: 2026-07-18
• Authors: Zi Wang, Caixia Yin, Siyuan Chen, Hao Zhang, Maoguang Chen, Tao Lin, Qiuxiang Tang
• DOI: 10.1016/j.agwat.2026.110608

Research Groups

College of Agronomy, Engineering Research Centre of Cotton, Xinjiang Agricultural University, Xinjiang 830052, China; Xinjiang Cotton Technology Innovation Center, Xinjiang Key Laboratory of Cotton Genetic Improvement and Intelligent Production, National Cotton Engineering Technology Research Center, Cotton Research Institute of Xinjiang Uyghur Autonomous Region Academy of Agricultural Sciences, Urumqi, Xinjiang 830091, China

Short Summary

This study developed a cotton plant moisture content (PMC) estimation model based on multimodal UAV remote sensing data. The CatBoost algorithm integrating multimodal data performed best, achieving R² and RMSE values of 0.807 and 0.033%, respectively.

Objective

• To identify key remote sensing features for characterizing changes in cotton PMC.
• To clarify the complementary role of multimodal remote sensing data.
• To determine suitable machine learning algorithms for multimodal data fusion modeling to construct an accurate PMC estimation model.

Study Configuration

• Spatial Scale: Field scale (110.4 m²).
• Temporal Scale: Cotton growth period from May to October 2024.

Methodology and Data

• Models used: CatBoost, AdaBoost, Random Forest, K-nearest neighbors.
• Data sources: UAV multimodal data (visible-light, multispectral, thermal infrared images), ground-measured PMC data.

Main Results

• The CatBoost model integrating multimodal data performed best, achieving R² and RMSE values of 0.807 and 0.033%, respectively.
• The suitable reference levels of PMC at different growth stages were identified based on the response relationship between measured PMC and yield under different water gradients.
• The results showed that the RFE feature selection method identified eight water-sensitive parameters.

Contributions

• This study realized high-precision remote sensing monitoring of PMC and further explored the appropriate moisture content thresholds for different growth stages, providing a quantitative reference for precision water regulation in cotton fields.

Funding

• National Cotton Engineering Technology Research Center (project code: 2020ZX04024001).
• Xinjiang Key Laboratory of Cotton Genetic Improvement and Intelligent Production (project code: XJLKYCXGCIP2020002).

Citation

@article{Wang2026Estimation,
  author = {Wang, Zi and Yin, Caixia and Chen, Siyuan and Zhang, Hao and Chen, Maoguang and Lin, Tao and Tang, Qiuxiang},
  title = {Estimation of cotton plant moisture content using UAV multimodal data and machine learning},
  journal = {Agricultural Water Management},
  year = {2026},
  doi = {10.1016/j.agwat.2026.110608},
  url = {https://doi.org/10.1016/j.agwat.2026.110608}
}