PDF
Full
Outline
To address the limitations of existing wheat biomass estimation models, such as insufficient generalizability in field environments and discontinuity in application due to growth-stage-specific modeling, this study focused on winter wheat from the jointing to filling stages in Henan Province. Multi-spectral UAV data were used to construct a 34-dimensional feature set, including reflectance from 10 spectral bands, 15 vegetation indices, 8 texture features, and quantified growth stages (jointing=1, booting=2, filling=3). Features were selected through correlation analysis (ranked by absolute correlation coefficient values), and four machine learning algorithms—Linear Regression (LR), Random Forest (RF), LightGBM, and K-Nearest Neighbors Regression (KNN)—were employed to develop a unified model for estimating above-ground biomass (AGB) across multiple growth stages. Model parameters were optimized by progressively increasing the number of input features. The results showed that the quantified growth stage had the highest correlation with AGB (correlation coefficient of 0.80), while the near-infrared band and the red-edge 740 nm band were identified as critical spectral features (with correlation coefficients of 0.48 and 0.44, respectively). The Random Forest (RF) model achieved the highest accuracy with 18 input features (growth stage + 2 band reflectance values + 8 vegetation indices + 7 texture features), with a coefficient of determination (R2) of 0.87, a root mean square error (RMSE) of 291.2 g·m-2, and a normalized root mean square error (nRMSE) of 12.8% on the test set. These findings demonstrate that a unified model for estimating AGB in winter wheat across multiple growth stages can be constructed using multi-spectral UAV data and field samples. This approach significantly enhances the model’s generalizability in practical field production environments and effectively addresses application challenges during transitional growth stages.
PDF
27
11
| 科 Family | 属数 Number of genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) | 属 Genus | 种数 Number of species | 占总种数比例 Percentage of total species (%) |
|---|---|---|---|---|---|---|
| 鹅膏菌科Amanitaceae | 2 | 11 | 5.26 | 鹅膏菌属 Amanita | 10 | 4.78 |
| 小菇科 Mycenaceae | 2 | 12 | 5.74 | 丝盖伞属 Inocybe | 5 | 2.39 |
| 多孔菌科 Polyporaceae | 8 | 14 | 6.70 | 蜡蘑属 Laccaria | 5 | 2.39 |
| 红菇科 Russulaceae | 3 | 23 | 11.00 | 小皮伞属 Marasmius | 6 | 2.87 |
| 小菇属 Mycena | 11 | 5.26 | ||||
| 光柄菇属 Pluteus | 5 | 2.39 | ||||
| 红菇属 Russula | 17 | 8.13 | ||||
| 栓菌属 Trametes | 5 | 2.39 |
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62199257
qkjq@cast.org.cn
Copyright © 2025 China Association for Science and Technology. All rights reserved. For all open access content, the relevant licensing terms apply.
Sponsored by the Office of the Leading Group for Cybersecurity and Informatization of CAST, and supported by Science and Technology Review Publishing House
