Remote Sensing Monitoring of Nitrogen Nutrient Index in Winter Wheat by Integrating Hyperspectral and Digital Imagery
YANG Fu-qin1, LI Chang-hao1, ZHANG Ying-fa1, CHEN Ri-qiang2, LIU Yang2, GUO Liang-dong3, FENG Hai-kuan2, 4*
1. College of Civil Engineering, Henan University of Engineering, Zhengzhou 451191, China
2. Key Laboratory of Quantitative Remote Sensing in Agriculture of Ministry of Agriculture and Rural Affairs, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China
3. Zhengzhou Shangpai Lanlian Network Technology Corporation, Zhengzhou 450000, China
4. National Engineering and Technology Center for Information Agriculture, Nanjing Agricultural University, Nanjing 210095, China
Abstract：Rapid, real-time, and accurate acquisition of the nitrogen nutrition status of winter wheat is crucial for evaluating winter wheat growth, estimating yield, and guiding agricultural modernization and production. This study utilized hyperspectral cameras and digital cameras mounted on drones to collect canopy spectral data during the three critical growth periods. Simultaneous ground experiments were conducted to determine the physical and chemical properties of biomass and plant nitrogen content. Four characteristic parameters, including the vegetation index, the red edge index, the red edge parameter, and the three-band parameter of the hyperspectral image, as well as the color index of the digital camera and its fusion parameters, were selected. Partial Least Squares Regression (PLSR), Stepwise Regression (SWR), Random Forest (RF), and Back Propagation (BP) algorithms were used to establish a winter wheat nitrogen nutrition index monitoring model. The accuracy of the model was evaluated, and the optimal estimation model was selected. The results showed that (1) In univariate modeling, the nitrogen nutrition index model constructed with the red-edge parameter DIDRmid was the best, achieving a modeling R2 of 0.66, RMSE of 0.11%, and a validation R2 of 0.55, RMSE of 0.13%. (2) In the multivariate modeling, the nitrogen nutrient index model constructed with the red edge parameter as the independent variable was superior to the nitrogen nutrient index model constructed with the vegetation index, the red edge index, the three-band parameter and the color index as the independent variables, where the nitrogen nutrition index model constructed by the BP algorithm based on red edge parameters was optimal (modeling R2=0.75, RMSE=0.10%, validation R2=0.60, RMSE=0.12%).(3) In fusing hyperspectral parameters and digital index variable modelling, the nitrogen nutrient index model constructed with the multimodal variable red edge parameter + color index was superior to the nitrogen nutrient index models constructed with the red edge parameter + vegetation index, red edge parameter + red edge index and red edge parameter + triple band parameter, where the nitrogen nutrient index model constructed using PLSR with the fusion of the red edge parameters and color index was optimal (modeling R2=0.77, RMSE=0.09%, validation R2=0.65, RMSE=0.11%). Its model accuracy was better than that of univariate modeling and multivariate modeling. This study can provide an important reference for estimating the nitrogen nutrition status of winter wheat.
Key words：Winter wheat; UAV; Nitrogen nutrient index; Spectral parameters; Color index
杨福芹，李昌浩，张英发，陈日强，刘 杨，郭良栋，冯海宽. 融合高光谱和数码影像的冬小麦氮营养指数遥感监测[J]. 光谱学与光谱分析, 2025, 45(06): 1719-1728.
YANG Fu-qin, LI Chang-hao, ZHANG Ying-fa, CHEN Ri-qiang, LIU Yang, GUO Liang-dong, FENG Hai-kuan. Remote Sensing Monitoring of Nitrogen Nutrient Index in Winter Wheat by Integrating Hyperspectral and Digital Imagery. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2025, 45(06): 1719-1728.
[1] YANG Fu-qin, DAI Hua-yang, FENG Hai-kuan, et al(杨福芹, 戴华阳, 冯海宽, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2016, 32(23): 161.
[2] Rodriguez I A, Lacasa J, van Versendaal E, et al. European Journal of Agronomy, 2024, 154: 127079.
[3] Cao Q, Miao Y, Feng G, et al. Computers and Electronics in Agriculture, 2015, 112: 54.
[4] WANG Yu-na, LI Fen-ling, WANG Wei-dong, et al(王玉娜, 李粉玲, 王伟东, 等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报), 2020, 36(22): 31.
[5] Yue S, Meng Q, Zhao R, et al. Agronomy Journal, 2012, 104(2): 523.
[6] Jiang J, Atkinson P M, Zhang J, et al. European Journal of Agronomy, 2022, 138: 126537.
[7] Han L, Yang G J, Dai H Y, et al. Plant Methods, 2019, 15: 10.
[8] Qiu Z, Ma F, Li Z, et al. Computers and Electronics in Agriculture, 2021, 189: 106421.
[9] Justes E, Mary B, Meynard J M, et al. Annals of Botany, 1994, 74(4): 397.
[10] Lemaire G, Jeuffroy M H, Gastal F. European Journal of Agronomy, 2008, 28(4): 614.
[11] Li F, Mistele B, Hu Y C, et al. Agricultural and Forest Meteorology, 2013, 180: 44.
[12] Huang S, Miao Y, Zhao G, et al. Remote Sensing, 2015, 7(8): 10646.
[13] Buschmann C, Nagel E. International Journal of Remote Sensing, 1993, 14(4): 711.
[14] Li L, Ren T, Ma Y, et al. Computers and Electronics in Agriculture, 2016, 126: 21.
[15] Guyot G, Baret F, Major D J. International Society of Photogrammetry and Remote Sensing, 1988, 27(11): 750.
[16] Cao Q, Miao Y, Wang H, et al. Field Crops Research, 2013, 154: 133.
[17] Huete A, Didan K, Miura T, et al. Remote Sensing of Environment, 2002, 83(1-2): 195.
[18] Broge N H, Leblanc E. Remote Sensing of Environment, 2001, 76(2): 156.
[19] HE Cai-lian, ZHENG Shun-lin, WAN Nian-xin, et al(何彩莲, 郑顺林, 万年鑫, 等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2016, 36(9): 2930.
[20] Wang D, Li R, Liu T, et al. Field Crops Research, 2023, 304: 109175.
