Accurate Estimation of Leaf Nitrogen Content in Broomcorn Millet Using Deep Learning, RGB Imagery, and Multi-Source Data Fusion
Agronomy · 15.08.2026 · 16 (16) · 1570
DOI: 10.3390/agronomy16161570 ↗Kurz gefasst
Fachartikel in der Zeitschrift Agronomy, online erschienen am 15.08.2026. Verfasst von Shike Zhao und 4 weiteren Autor:innen. Die Arbeit nennt 67 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Shanxi Agricultural University Special and Excellent Engineering Project, Shanxi Province Basic Research Project, Shanxi Agricultural University “Biological Breeding Engineering” Project.
Abstract (Originalsprache)
Nitrogen is a major determinant of crop growth, yield formation, and grain quality, making precise nitrogen (N) management essential for sustainable production. This study integrated field canopy spectra acquired with an ASD FieldSpec 4 spectrometer and unmanned aerial vehicle (UAV)-based RGB imagery to estimate leaf nitrogen content in broomcorn millet across four growth stages. Vegetation indices (VIs) and gray-level co-occurrence matrix texture features were derived, and six algorithms—partial least squares (PLS), support vector machine (SVM), random forest (RF), one-dimensional convolutional neural network (CNN1D), one-dimensional residual network (ResNet1D), and one-dimensional U-Net (U-Net)—were evaluated. Among the spectral preprocessing methods, the first-derivative transformation showed the strongest relationship with leaf nitrogen content (r = −0.85). Models based on multi-source feature fusion consistently outperformed those using a single data source. The U-Net model using ASD + RGB + VIs + texture features achieved the highest test-set accuracy (R2 = 0.913, RMSE = 1.851, and RPD = 3.382). SHapley Additive exPlanations (SHAP) analysis identified blue-band correlation (B_Correlation) as the most influential individual predictor (mean absolute SHAP value = 0.39), while texture features accounted for 45.170% of the cumulative feature importance. These results demonstrate the potential of multi-source remote sensing and deep learning for rapid, non-destructive assessment of leaf nitrogen content and precision nitrogen management in broomcorn millet.
Förderung
- Shanxi Agricultural University Special and Excellent Engineering Project
- Shanxi Province Basic Research Project
- Shanxi Agricultural University “Biological Breeding Engineering” Project
Zitieren
APA
Zhao, S., Chang, B., Zhang, Y., Qiao, Z., & Wang, J. (2026). Accurate Estimation of Leaf Nitrogen Content in Broomcorn Millet Using Deep Learning, RGB Imagery, and Multi-Source Data Fusion. Agronomy, 16(16), 1570. https://doi.org/10.3390/agronomy16161570
Vancouver
Zhao S, Chang B, Zhang Y, Qiao Z, Wang J. Accurate Estimation of Leaf Nitrogen Content in Broomcorn Millet Using Deep Learning, RGB Imagery, and Multi-Source Data Fusion. Agronomy. 2026;16(16):1570. doi:10.3390/agronomy16161570
BibTeX
@article{Zhao2026,
author = {Zhao, Shike and Chang, Bo and Zhang, Yiting and Qiao, Zhijun and Wang, Junjie},
title = {Accurate Estimation of Leaf Nitrogen Content in Broomcorn Millet Using Deep Learning, RGB Imagery, and Multi-Source Data Fusion},
journal = {Agronomy},
year = {2026},
volume = {16},
number = {16},
pages = {1570},
doi = {10.3390/agronomy16161570}
}
EUVIMED