作物学报2025，Vol.51Issue(5)：1326-1337,12.DOI:10.3724/SP.J.1006.2025.44157
基于高光谱遥感的油菜叶片氮磷养分含量诊断
Diagnosis of nitrogen and phosphorus nutrient content in rapeseed leaves based on hyperspectral remote sensing
摘要
Abstract
Hyperspectral remote sensing technology provides an accurate and non-destructive method for diagnosing nitrogen(N)and phosphorus(P)deficiencies in rapeseed,laying the groundwork for precision fertilization.This study utilized multi-site,multi-year field trials to collect data on leaf nitrogen concentration(LNC),leaf phosphorus concentration(LPC),yield,and the canopy reflectance spectrum of winter rapeseed during the overwintering period.Feature bands sensitive to LNC and LPC were identified using competitive adaptive reweighted sampling(CARS),successive projections algorithm(SPA),and the elimination of non-informative variables(UVE).Partial least squares regression(PLSR)models were constructed to estimate LNC and LPC based on both the original spectrum and the first-order derivative spectrum.Nutrient deficiency diagnosis was achieved by inte-grating the nitrogen nutrition index(NNI)and phosphorus nutrition index(PNI)derived from the estimated nutrient concentra-tions.The results revealed that the characteristic bands for LNC and LPC were primarily concentrated in the ranges of 400-460 nm,650-730 nm,1140-1210 nm,and 2240-2370 nm for LNC,and 650-730 nm,2100-2310 nm for LPC.The model based on the first-order derivative spectrum and the UVE method demonstrated superior accuracy compared to other models.In the test set,the model achieved high estimation accuracy for LNC(R2=0.773,RMSE=0.528%)and LPC(R2=0.785,RMSE=0.09%).Thresh-old values for NNI and PNI during the overwintering period were established using yield data from field trials,which were 1.20 and 0.75,respectively.By employing hyperspectral remote sensing to estimate LNC and LPC,subsequent calculations of NNI and PNI can effectively diagnose nutrient deficiencies in rapeseed during the overwintering period.This approach provides a novel technological solution for the sustainable development of rapeseed production.关键词
冬油菜/高光谱遥感/偏最小二乘/波段选择/叶片氮磷含量Key words
winter rape/hyperspectral remote sensing/partial least squares/band selection/leaf nitrogen and phosphorus concen-tration引用本文复制引用
王清华,朱格格,方雯,刘诗诗,鲁剑巍..基于高光谱遥感的油菜叶片氮磷养分含量诊断[J].作物学报,2025,51(5):1326-1337,12.基金项目
本研究由国家自然科学基金项目(42171350)和国家重点研发计划项目(2021YFD1600503)资助. This study was supported by the National Natural Science Foundation of China(42171350)and the National Key Research and Development Program of China(2021YFD1600503). （42171350）
