UAV-based spatiotemporal phenotyping and growth modeling for forecasting potato yield

Monitoring spatial variations in plant growth and forecasting yield before harvest provides valuable insights for optimizing agronomic decision-making in potato cultivation. Although unmanned aerial vehicle (UAV)-based remote sensing has recently enabled the development of tuber fresh weight (TW) estimation models, their integration into practical yield-forecasting systems remains limited. In this study, we developed machine learning models to estimate tuber weights at multiple preharvest time points using RGB and multispectral UAV imagery. Image-derived features were extracted from the orthomosaic and digital surface model (DSM) images for each plot, and a random forest regression model was trained for TW estimation. The estimated values were subsequently used to fit the Gompertz growth curves, which were then used to forecast the yield at the expected harvest time. The correlation between the estimated and observed values was strong in the UAV-based TW estimation, with correlation coe..., , # Data from: UAV-based spatiotemporal phenotyping and growth modeling for forecasting potato yield [Access this dataset on Dryad](https://doi.org/10.5061/dryad.5qfttdzmn) This dataset provides the data and reproducible code used for pre-harvest yield forecasting of potato using UAV-based remote sensing. Field experiments were conducted in 2023 and 2024. For each experimental plot, RGB and multispectral UAV imagery were acquired, and vegetation features were extracted from orthomosaic images and digital surface models (DSM). In addition, destructive sampling was conducted at multiple time points to measure tuber fresh weight (TW). Using these UAV-derived features and observed TW data, a Random Forest regression model was developed to estimate TW at different harvest stages. Furthermore, Gompertz growth curves were fitted to the time-series of estimated TW in order to forecast yield at the expected harvest date. This repository includes plot-level ground-truth TW datasets (.csv), UAV-..., ,监测作物生长的空间变异并在收获前预测产量，可为优化马铃薯栽培的农艺决策提供重要参考。尽管基于无人机（Unmanned Aerial Vehicle, UAV）的遥感技术近年已推动块茎鲜重（Tuber Fresh Weight, TW）估算模型的研发，但此类模型在实际产量预测系统中的落地应用仍较为有限。本研究借助RGB与多光谱无人机影像，构建了可在多个收获前时间节点估算块茎重量的机器学习模型。针对每个试验小区，从正射影像与数字表面模型（Digital Surface Model, DSM）中提取影像特征，并训练随机森林回归模型以实现块茎鲜重估算。随后将估算得到的块茎重量值用于拟合Gompertz生长曲线，进而预测目标收获期的总产量。基于无人机的块茎鲜重估算中，估算值与实测值间相关性极强，相关系数……，，# 数据来源：基于无人机时空表型分析与生长建模的马铃薯产量预测 [在Dryad平台获取本数据集](https://doi.org/10.5061/dryad.5qfttdzmn) 本数据集包含了利用基于无人机的遥感技术开展马铃薯收获前产量预测所使用的实验数据与可复现代码。试验分别于2023年与2024年开展。针对每个试验小区，采集RGB与多光谱无人机影像，并从正射影像及数字表面模型中提取植被特征。此外，在多个时间节点开展破坏性采样以测定块茎鲜重（TW）。基于上述无人机影像提取的特征与实测块茎鲜重数据，构建随机森林回归模型以估算不同收获阶段的块茎重量。进一步通过拟合估算块茎重量的时间序列数据得到Gompertz生长曲线，从而预测目标收获期的总产量。本数据集仓库包含小区级实测块茎鲜重数据集（.csv格式）、无人机影像相关数据……，，