Monitoring of Nitrogen Content in Winter Wheat Based on UAV
Hyperspectral Imagery
FENG Hai-kuan1, 2, FAN Yi-guang1, TAO Hui-lin1, YANG Fu-qin3, YANG Gui-jun1, ZHAO Chun-jiang1, 2*
1. Key Laboratory of Quantitative Remote Sensing in Agriculture, Ministry of Agriculture, Information Technology Research Center, Beijing Academy of Agriculture and Forestry Sciences, Beijing 100097, China
2. National Engineering and Technology Center for Information Agriculture, Nanjing Agricultural University, Nanjing 210095, China
3. College of Civil Engineering, Henan University of Engineering, Zhengzhou 451191, China
Abstract:The nitrogen content of crops affects the growth status of crops. A suitable nitrogen content can greatly improve the growth and yield of crops. Therefore, it is very important to monitor nitrogen content quickly. This study aimed to explore the potential of combining vegetation indices and spectral feature parameters acquired by UAV imaging hyperspectral to improve the accuracy of nitrogen content estimation during key growth stages of winter wheat. Firstly, the UAV was used as a remote sensing platform with hyperspectral sensors to acquire hyperspectral remote sensing images of four major growth stages of winter wheat: plucking, flag picking, flowering, and filling stages, and the nitrogen content data of each growth stage were measured. Secondly, based on pre-processed hyperspectral images, we extracted the canopy reflectance data of winter wheat at each growth stage. As a result, we constructed 12 vegetation indices and 12 spectral feature parameters that can better reflect the nitrogen nutrient status of the crop. Then, the correlation between the spectral parameters and the nitrogen content of winter wheat was calculated, and vegetation indices and spectral feature parameters with a strong correlation with the nitrogen content in each growth period were screened out. Finally, a nitrogen content estimation model based on vegetation indices and vegetation indices combined with spectral feature parameters was constructed using Stepwise Regression (SWR) analysis. The results showed that (1) most of the selected vegetation indices and spectral feature parameters were highly correlated with the N content of winter wheat. Among them, the correlation of vegetation indices was higher than that of spectral feature parameters; (2) although it is feasible to estimate winter wheat based on individual vegetation indices or spectral feature parameters, the accuracy needs to be further improved. (3) compared with a single vegetation index or spectral feature parameter, the accuracy and stability of the nitrogen content estimation model constructed by vegetation index combined with spectral feature variables using the SWR method were higher (at the plucking stage: modeling R2=0.64, RMSE=24.68%, NRMSE=7.96%, validation R2=0.77, RMSE=23.13%, NRMSE=7.81%; flag picking phase: modeling R2=0.81, RMSE=15.79%, NRMSE=7.41%, validation R2=0.84, RMSE=15.10%, NRMSE=7.08%; flowering phase: modeling R2=0.78, RMSE=9.88%, NRMSE=5.66%, validation R2=0.85, RMSE=9.12%, NRMSE=4.76%; filling stage: modeling R2=0.49, RMSE=13.68%, NRMSE=9.85%, validation R2=0.40, RMSE=18.29%, NRMSE=14.73%). The results showed high accuracy and stability of the winter wheat N content estimation model constructed by combining vegetation indices and spectral feature parameters obtained by UAV imaging hyperspectral. The research results can provide a reference for the spatial distribution and precise management of winter wheat N content.
冯海宽,樊意广,陶惠林,杨福芹,杨贵军,赵春江. 利用无人机高光谱影像的冬小麦氮含量监测[J]. 光谱学与光谱分析, 2023, 43(10): 3239-3246.
FENG Hai-kuan, FAN Yi-guang, TAO Hui-lin, YANG Fu-qin, YANG Gui-jun, ZHAO Chun-jiang. Monitoring of Nitrogen Content in Winter Wheat Based on UAV
Hyperspectral Imagery. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2023, 43(10): 3239-3246.
[1] He L, Song X, Feng W, et al. Remote Sensing of Environment, 2016, 174: 122.
[2] Bareth G, Aasen H, Bendig J, et al. Photogrammetrie Fernerkundung Geoinformation,2015,(1): 69.
[3] LIU Shuai-bing, YANG Gui-jun, JIN Hai-tao, et al(刘帅兵,杨贵军,景海涛,等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报),2019, 35(11): 75.
[4] JING Yu-hang, GUO Yan, ZHANG Hui-fang, et al(井宇航,郭 燕,张会芳,等). Journal of Henan Agricultural Sciences(河南农业科学), 2022, 51(2): 147.
[5] Rodene E, XU G, Delen S P, et al. The Plant Phenome Journal, 2022, 5(1): e20030.
[6] FAN Yi-guang, FENG Hai-kuan, LIU Yang,et al(樊意广,冯海宽,刘 杨,等). Transactions of the Chinese Society for Agricultural Machinery(农业机械学报), 2022, 53(6): 202.
[7] Wang W, Wu Y, Zhang Q, et al. IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2021, 14: 6716.
[8] FAN Yi-guang, FENG Hai-kuan, LIU Yang, et al(樊意广,冯海宽,刘 杨,等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2022, 42(10): 3217.
[9] Tao H, Feng H, Xu L, et al. Sensors, 2020, 20(4):1231.
[10] WU Wei-bin, LI Jia-yu, ZHANG Zhen-bang, et al(吴伟斌,李佳雨,张震邦,等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报),2018, 34(3): 195.
[11] WEI Peng-fei, XU Xin-gang, LI Zhong-yuan, et al(魏鹏飞,徐新刚,李中原,等). Transactions of the Chinese Society of Agricultural Engineering(农业工程学报),2019, 38(5): 126.