【摘要】 快速准确地获取葡萄关键生育期的元素含量，是现代精准农业的基本要求。利用无人机搭载五通道多光谱相机，采集4个葡萄关键生育期多光谱影像数据，并以相应实测葡萄叶片氮含量为基准，进行葡萄叶片氮含量估测遥感无损监测研究。通过皮尔逊相关性分析，筛选出16个与氮含量强相关的光谱变量，并采用后退逐步回归分析法逐个剔除不显著变量，综合考虑模型精度与简洁度，在不同生育期选择合适的光谱变量进行建模，实现对葡萄叶片氮含量较高精度检测。在各生育期选取无人机多光谱遥感最优光谱变量监测葡萄氮素含量，为葡萄园的精准管理提供决策信息和技术支持。更多还原

【Abstract】 Rapid and accurate acquisition of element content in main grape growth periodsis the basic requirement of modern precision agriculture.a UAV-mounted five-channel multi-spectral camera was used to collect multi-spectral image data from four key growth periods of grapes, and based on the corresponding measured grape leaf nitrogen content as a benchmark, research on the remote sensing non-destructive monitoring of grape leaf nitrogen content estimation was conducted.Through the Pearson correlation analysis, 16 spectral variables strongly correlated with nitrogen content were screened out,the insignificant variables were eliminated one by one with the stepwise regression analysismethod. Appropriate spectral variables were selected for modeling in different growth periods considering the model accuracy and simplicity,realizing the high-precision detection of nitrogen content in grape leaves. In each growth period,the optimal spectral variables of UAV multi-spectral remote sensing were selected to monitor the nitrogen content of grapes,so as to provide decision-making information and technical support for the precise management of vineyards.更多还原