Study on Hyperspectral Estimation Model of Total Nitrogen Content in Soil of Shaanxi Province

2017 3rd International Conference on Environmental Science and Material Application

生态环境科学;材料科学

Liu, Jinbao^1,3,4,5 ; Dong, Zhenyu^2 ; Chen, Xi^1,3,4,5

Institute of Land Engineering and Technology, Shaanxi Provincial Land Engineering Construction Group Co., Ltd., Xi'an

710075, China^1

College of Forestry, Shenyang Agricultural University, Shenyang

110866, China^2

Key Laboratory of Degraded and Unused Land Consolidation Engineering, Ministry of Land and Resources, Xi'an

710075, China^3

Shaanxi Provincial Land Engineering Construction Group Co., Ltd, Xi'an

710075, China^4

Shaanxi Provincial Land Consolidation Engineering Technology Research Center, Xi'an

710075, China^5

关键词: Correlation coefficient;    First-order differentials;    Hyperspectral remote sensing technology;    Least-squares regression method;    Logarithmic transformations;    Principal component regression;    Reflectance spectroscopy;    Total nitrogen content;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/108/4/042025/pdf
DOI  :  10.1088/1755-1315/108/4/042025

来源: IOP

【 摘 要 】

The development of hyperspectral remote sensing technology has been widely used in soil nutrient prediction. The soil is the representative soil type in Shaanxi Province. In this study, the soil total nitrogen content in Shaanxi soil was used as the research target, and the soil samples were measured by reflectance spectroscopy using ASD method. Pre-treatment, the first order differential, second order differential and reflectance logarithmic transformation of the reflected spectrum after pre-treatment, and the hyperspectral estimation model is established by using the least squares regression method and the principal component regression method. The results show that the correlation between the reflectance spectrum and the total nitrogen content of the soil is significantly improved. The correlation coefficient between the original reflectance and soil total nitrogen content is in the range of 350 ∼ 2500nm. The correlation coefficient of soil total nitrogen content and first deviation of reflectance is more than 0.5 at 142nm, 1963nm, 2204nm and 2307nm, the second deviation has a significant positive correlation at 1114nm, 1470nm, 1967nm, 2372nm and 2402nm, respectively. After the reciprocal logarithmic transformation of the reflectance with the total nitrogen content of the correlation analysis found that the effect is not obvious. Rc2= 0.7102, RMSEC = 0.0788; Rv2= 0.8480, RMSEP = 0.0663, which can achieve the rapid prediction of the total nitrogen content in the region. The results show that the principal component regression model is the best.