Inversion modeling of rice canopy nitrogen content based on MPA-GA-ELM UAV hyperspectral remote sensing
Abstract
Abstract: Nitrogen content is an important indicator to characterize the growth status of rice. UAV hyperspectral remote sensing technology was used to obtain the nitrogen content of rice canopy at regional scale in a timely manner. To improve the accuracy of high-spectral inversion of rice canopy nitrogen content, the marine predators algorithm (MPA) was used to downscale the hyperspectral information in the range of 400 nm to 1000 nm by using the UAV hyperspectral image data and the synchronously measured rice canopy nitrogen content as the data source, from which the hyperspectral characteristic variables for rice nitrogen content inversion modelling were extracted. And using the dimensionality reduction variables as the data base, two neural network inversion methods, including extreme learning machines (ELM), genetic algorithm optimization for extreme learning machines (GA-ELM), were used to establish the rice nitrogen content drone hyperspectral remote sensing inversion model, and the results showed that: (1) The hyperspectral range from 400 nm to 1000 nm was dimensionally reduced by MPA, and finally the continuous hyperspectral reflectance information was dimensionally reduced to four discrete hyperspectral feature wavelengths, 570, 723, 811 and 987 nm for subsequent inversion modeling of rice nitrogen content. (2) In the models adopted in this study, MPA-GA-ELM has the highest accuracy, where the R2 of training data, the R2 of test data, the RMSE of training data, and the RMSE of test data were 07984, 0.7357, 0.4615, 0.4878, respectively. This study provides data support and application basis for inverse UAV remote sensing diagnosis of nitrogen content in rice.
Keywords: UAV, Nitrogen inversion, rice, MPA, ELM, hyperspectral image
DOI: 10.33440/j.ijpaa.20210402.173
Citation: Yu F H, Guo Z H, Xu T Y. Inversion modeling of rice canopy nitrogen content based on MPA-GA-ELM UAV hyperspectral remote sensing. Int J Precis Agric Aviat, 2021; 4(2): 30–35.
Full Text:
PDFReferences
Xu B, Xu T Y, Yu F H, Zhang G S, Feng S, Guo Z H, Zhou C X. Near infrared spectral inversion of cellulose content in rice stems in Northeast Cold Area. Spectroscopy and Spectral Analysis, 2021,41(06): 1775–1781.
Du W, Xu T, Yu F, et al. Measurement of nitrogen content in rice by inversion of hyperspectral reflectance data from an un manned aerial vehicle. Ciência Rural, 2018,48(6). doi: 10.1590/0103-8478cr20180008
Qiuxiang Y I, Huang J, Wang F, et al. Evaluating the performance of PC-ANN for the estimation of rice nitrogen concentration from canopy hyperspectral reflectance. International Journal of Remote Sensing, 2010, 31(4): 931–940. doi: 10.1080/0143116 0902912061
Yu F H, Feng S, Zhao S, Wang D K, Xing S, Xu T Y. Estimation of chlorophyll content in Japonica Rice Canopy by pso-elm hyperspectral remote sensing inversion. Journal of South China Agricultural University, 2020,41(06): 59–66.
Yu F , Feng S , Du W , et al. A study of nitrogen deficiency inversion in rice leaves based on the hyperspectral reflectance differential. Frontiers in Plant Science, 2020, 11. doi: 10.3389/fpls.2020.573272
Xue L H, Luo W H, Cao W X, Tian Y C. Progress in spectral diagnosis of crop water and nitrogen. Journal of rRemote Sensing, 2003, (01): 73–80.
Zhao S. Study on Hyperspectral rapid detection method of chlorophyll content in rice canopy in cold area. Shenyang Agricultural University, 2020.
Kokaly R F, Clark R N.1999.Spectroscopic determination of leaf biochemistry using band-depth analysis of absorption features and stepwise multiple linear regression. Remote Sensing of Environment, 67: 267–287. doi: 10.1016/S0034-4257(98)00084-4.
Tian T, Zhang Q, Zhang H D. Research progress on the application of UAV Remote Sensing in crop monitoring. The Crop Journal, 2020, (05): 1–8.
Li X Q, li L, Zhuang L Y, Liu W Q, Liu X G, Li J. Inversion of heavy metal content in rice canopy based on wavelet transform and BP neural network. Transactions of the CSAM, 2019, 50(06): 226–232.
Zhu C, Miao T, Xu T Y, Li N, Deng H B, Zhou Y C. 3D point cloud ear segmentation and phenotypic parameter extraction of maize plant based on skeleton. Transactions of the CSAE, 2021, 37(06): 295–301.
Tan K, Wang S, Song Y, et al. Estimating nitrogen status of rice canopy using hyperspectral reflectance combined with BPSO-SVR in cold region. Chemometrics and Intelligent Laboratory Systems, 2017, 172:68–79. doi: 10.1016/j.chemolab.2017.11.014
Moharana S, Dutta S. Hyperspectral remote sensing of paddy crop using insitu measurement and clustering technique. ISPRS - International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 2014, XL-8(8): 845–851. doi: 10.5194/isprsarchives-XL-8-845- 2014.
Xu T Y, Hu K Y, Zhou Y C, Yu F H, Feng S. Extraction method of Japonica Rice from Landsat 8 image based on cart decision tree and BP neural network. Journal of Shenyang Agricultural University, 2020, 51(02): 169–176.
Dunn B, Dehaan R, Schmidtke L, et al. Using field-derived hyperspectral reflectance measurement to identify the essential wavelengths for predicting nitrogen uptake of rice at panicle initiation. Journal of Near Infrared Spectroscopy, 2016, 24(5): 473.
Yu F H, Xu T Z, Guo Z H, Du W, Wang D K, Cao Y L. Remote sensing inversion of chlorophyll content in cold rice leaves based on red edge optimized vegetation index. Smart Agriculture (Chinese and English), 2020, 2(01): 77–86.
Wang W, Yao x, Tian Y C, et al. Common Spectral Bands and Optimum Vegetation Indices for Monitoring Leaf Nitrogen Accumulation in Rice and Wheat. Journal of Agricultural Sciences: English edition, 2012 (issue 12): 2001–2012 doi: 10.1016/S2095-3119(12)60457-2
Xu T Y, Guo Z H, Yu F H, Xu B, Feng S. Diagnosis method of nitrogen deficiency of rice in cold area using ga-elm. Transactions of the CSAE, 2020, 36(02): 209–218.
Refbacks
- There are currently no refbacks.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.