|
|
|
|
|
|
| Predicting Soil Available Nitrogen with Field Spectra Corrected by Y-Gradient General Least Square Weighting |
| QI Hai-jun1, 2, Karnieli Arnon2, LI Shao-wen1* |
1. School of Information and Computer, Anhui Agricultural University, Hefei 230036, China
2. The Remote Sensing Laboratory, Jacob Blaustein Institutes for Desert Research, Ben-Gurion University of the Negev, Sede Boger Campus 84990, Israel |
|
|
|
|
Abstract Soil available nitrogen is supposed to be an important nutrient constituent for the growth and development of crops. In-situ field visible-near infrared (VIS-NIR, 350~2 500 nm) spectroscopic analysis is a rapid and non-destructive method that has the potential to predict nitrogen. Further, it is cost-effective method compared with traditional laboratory analysis and can be used to provide a database for the development of real-time soil nutrient sensors. However, prediction accuracy was greatly reduced due to unexpected environmental factors under field condition. In the current research, field work contained 76 samples from two sites located in the center and north parts of Israel. Y-gradient general least squares weighting (Y-GLSW) algorithm was investigated to filtering correct the field VIS-NIR spectra for improving the prediction ability of nitrogen. Firstly, Savitzky-Golay (SG) smoothing algorithm, first derivative transformation and standard normal variate were sequentially conducted to preprocess and transform the raw field spectra (RS). Then, a filtering model was established based on the Y-GLSW algorithm to correct the preprocessed and transformed spectra (PPT). After that, partial least square - regression (PLS-R) algorithm was applied to build regression models with RS, PPT, and Y-GLSW corrected spectra, respectively. As a result, the regression model based on RS was proved to be unfeasible. The ratio of performance to deviation (RPD) and the ratio between interpretable sum squared deviation and real sum squared deviation (SSR/SST) of the test set of the PPT-based regression model were found to be 1.41 and 0.57, respectively. The results of Y-GLSW-based regression model were RPD = 2.07 and SSR/SST=0.69 that significantly increased by 46.81% and 21.05% compared with PPT-based regression model. The results indicated that Y-GLSW was suitable to remove some unexpected variations (like the effect of environmental factors) of field spectra and improved the prediction accuracy and explanation ability of PLS-R model for predicting nitrogen.
|
|
Received: 2017-02-28
Accepted: 2017-07-05
|
|
|
|
Corresponding Authors:
LI Shao-wen
E-mail: shwli@ahau.edu.cn
|
|
[1] Paz-Kagan T, Shachak M, Zaady E, et al. Geoderma, 2014, s230-231(7): 171.
[2] WU Qian, YANG Yu-hong, XU Zhao-li, et al(吴 茜，杨宇虹，徐照丽，等). Spectroscopy and Spectral Analysis(光谱学与光谱分析), 2014, 34(8): 2102.
[3] Shao Y, He Y. Soil Research, 2011, 49(2): 166.
[4] Ji W, Shi Z, Huang J, et al. PLoS ONE, 2014, 9(8): e105708.
[5] Kodaira M, Shibusawa S. Geoderma, 2013, 199(4): 64.
[6] Liu X, Guo Y, Wang Q L, et al. Proc. SPIE, 2013, 8910: 891020(doi: 10.1117/12.2035010).
[7] Ji W, Viscarra Rossel R A, Shi Z. European Journal of Soil Science, 2015, 66(3): 555.
[8] Wold S, Antti H, Lindgren F, et al. Chemometrics and Intelligent Laboratory Systems，1998, 44(1): 175.
[9] Roger J M, Chauchard F, Bellon-Maurel V. Chemometrics and Intelligent Laboratory Systems, 2003, 66(2): 191.
[10] Martens H, Hy M, Wise B M, et al. Journal of Chemometrics, 2003, 17(3): 153.
[11] Liu Y, Jiang Q, Shi T, et al. Acta Agriculturae Scandinavica Section B-Soil and Plant Science, 2014, 64(3): 267.
[12] Jiang Q, Chen Y, Guo L, et al. Remote Sensing, 2016, 8(9): 755.
[13] Paz-Kagan T, Zaady E, Salbach C, et al. Remote Sensing, 2015, 7(11): 15748.
[14] Zorzetti B M, Shaver J M, Harynuk J J. Analytica Chimica Acta, 2011, 694(1): 31.
[15] Zhong P, Xu Y, Zhao Y. Neural Computing and Applications, 2012, 21(2): 399. |
| [1] |
ZHAO Xu-ting1, 2, 3, ZHANG Shu-juan1, LI Bin2, 3, 4*, LI Yin-kun5. Study on Moisture Content of Soybean Canopy Leaves under Drought Stress Using Terahertz Technology[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2018, 38(08): 2350-2354. |
| [2] |
ZHANG Hong-guang, LU Jian-gang* . An Improved ELM Algorithm for Near Infrared Spectral Quantitative Analysis[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2016, 36(09): 2784-2788. |
| [3] |
LI Bin, ZHANG Tao, JIA Er-hui . STR Fluorescence Spectral Analysis Based on Matrix Analysis [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2015, 35(02): 429-432. |
| [4] |
LU Yu, LI Xiang-ru*, YANG Tan, WANG Yong-jun . A New Method of Sparse Feature Extraction for Stellar Spectra [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2014, 34(08): 2279-2283. |
| [5] |
LIU Pei-zhong1, ZHANG Li-ping2*, LI Wei-jun2, QIN Hong2, DONG Xiao-li2 . Study on Spectral Calibration of Discrimination of Corn Variety Using Near-Infrared Spectra Based on DS Algorithm [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2014, 34(06): 1533-1537. |
| [6] |
CHEN Xiang-dong, GAO Feng, WU Ben-ke, CHENG Ping, YANG Ji-ping* . Investigations of Fluorescence Spectral Correction for a Two-Component Mixed Solution with Overlapping Absorption Spectra [J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2013, 33(03): 659-662. |
| [7] |
WU Hua-yong1, ZHOU Ze-yu1, WANG Hong-tao1*, LU Wen-jing1, SUN Xiao-jie2 . Effect of Spectra Correction on the Fluorescence Characteristics of Dissolved Organic Matter[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2012, 32(11): 3044-3048. |
| [8] |
SONG Tao,BAO Yi-dan*,HE Yong. Research on the Method for Rapid Detection of Soil Moisture Content Using Spectral Data[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2009, 29(03): 675-677. |
| [9] |
GAO Xiao-feng,XU Zhi-hai*. Comparison of Two Methods to Solve AR Model Parameters of Line Shape Optimized Maximum Entropy Spectral Estimation[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2008, 28(11): 2717-2720. |
| [10] |
AN Xin1,SU Shi-guang,WANG Tao1,XU Shuo1,HUANG Wen-jiang2,ZHANG Lu-da1* . Compound Support Vector Machines Method with Application in Spctral Analysis[J]. SPECTROSCOPY AND SPECTRAL ANALYSIS, 2007, 27(08): 1619-1621. |
|
|
|
|
