A spectral index for estimating grain filling rate of winter wheat using UAV-based hyperspectral images

文献类型： 外文期刊

第一作者： Zhang, Baoyuan

作者： Zhang, Baoyuan;Wu, Wenbiao;Zhou, Jingping;Dai, Menglei;Sun, Qian;Sun, Xuguang;Gu, Xiaohe;Zhang, Baoyuan;Dai, Menglei;Sun, Xuguang;Chen, Zhen

作者机构：

关键词： Grain filling rate; Thousand grain weight; UAV-based hyperspectral imaging; Winter wheat; Spectral index

期刊名称：COMPUTERS AND ELECTRONICS IN AGRICULTURE （ 影响因子：7.7； 五年影响因子：8.4 ）

ISSN： 0168-1699

年卷期： 2024 年 223 卷

页码：

收录情况： SCI

摘要： The grain filling rate (GFR) significantly affects grain weight and quality, playing a pivotal role in the winter wheat production. Traditional observation methods for grain filling rates rely on manual field sampling, which is both costly and time-consuming. This study aims to develop a spectral index using UAV-based hyperspectral image to estimate the grain filling rate of winter wheat analyze its dynamic changes. Initially, the savitzky-golay (SG) filtering algorithm was applied to smooth the original UAV hyperspectral data. Subsequently, according the contribution calculated by successive projections algorithm (SPA) and random forest algorithm (RF), three most sensitive bands of GFR were selected. Finally, the three selected sensitive bands were used to construct the grain filling rate spectral index (GFRSI) based on the angles formed on the spectral curve. The performance of GFRSI was compared with common spectral indices in the estimation of GFR and thousand-grain weight (TGW). The results indicate that GFRSI outperforms commonly used vegetation indices, achieving R2, RMSE, and NRMSE values of 0.79, 0.34 mg/d, and 12.5 %, respectively, for GFR estimation. Additionally, when considering the estimated GFR and grain filling duration (GFD), TGW estimation results exhibit R2, RMSE, and NRMSE values of 0.90, 4.66 g, and 9.0 %, respectively. Using the proposed model, we successfully achieved field-scale spatial mapping of GFR and TGW for winter wheat. The spectral index for grain filling rate constructed in this study provides valuable support and reference for the rapid estimation of winter wheat grain filling rates using UAV hyperspectral imaging technology.

分类号：

  • 相关文献
作者其他论文 更多>>

微信小程序