Soil moisture and salt information combined extraction method based on hyperspectral data

Abstract

The invention discloses a soil moisture and salt information combined extraction method based on hyperspectral data. The method includes the following steps: carrying out primary treatment of remote sensing hyperspectral reflectivity data, and according to the reflectivity, calculating the normalized reflectivity, an apparent absorption rate first-order derivative and an apparent absorption rate second-order derivative; establishing an evaluation system based on principal component analysis, and selecting sensitive wave bands; determining a first inversion variable; with use of the sensitive wave bands, using a stepwise regression method, establishing an inversion model of soil moisture, and partitioning the soil samples according to the predicted soil moisture content, to obtain soil sample intervals; and with the sensitive wave bands and the soil sample intervals, with use of the stepwise regression method, respectively establishing an inversion model of the soil salt content for the soil sample in each interval. The method has the advantages of improving the extraction accuracy of the soil salt.

Description

technical field [0001] The invention relates to the field of agriculture, more specifically it is a joint extraction method of soil moisture and salinity information based on hyperspectral data. Background technique [0002] Accurate acquisition of soil moisture and salt content plays an important role in agricultural development and ecological protection in arid and semi-arid areas; the traditional method of extracting soil water and salt is to collect samples in the field and bring them back to the laboratory for testing. This method has high precision. However, labor costs are high and time-consuming; the use of hyperspectral remote sensing data to extract large-area soil moisture and salinity information is an important scientific and technological support; however, the general method of extracting soil moisture does not consider the influence of soil salinity, and is not suitable for use in saline soils , because soil moisture and salinity have a comprehensive impact on...

AI Technical Summary

Problems solved by technology

[0002] Accurate acquisition of soil moisture and salt content plays an important role in agricultural development and ecological protection in arid and semi-arid areas; the traditional method of extracting soil water and salt is to collect samples in the field and bring them back to the laboratory for testing. This method has high precision. However, labor costs are high and time-consuming; the use of hyperspectral remote sensing data to extract large-area soil moisture and salinity information is an important scientific and technological support; however, the general method of extracting soil moisture does not consider the influence of soil salinity, and is not suitable for use in saline soils , because soil moisture and salinity have a comprehensive impact on soil spectra; similarly, the general method of extracting soil salinity does not consider the influence of soil moisture; therefore, it is difficult to extract soil moisture and salinity information in saline soils using hyperspectral data; at the same time, Studies have shown that when the soil salinity content is low, inversion is difficult; the salinity content in fields suitable for crop growth in arid and semi-arid areas is generally less than 1%, and the general inversion model is not applicable

[0003] Most of the currently used satellite hyperspectral data or hyperspectral data collected by ground object spectrometers are reflectance data; the reflectance data of visible light, near-infrared and short-wave infrared spectra are related to soil moisture, salinity, organic matter, color, iron oxide, Many properties such as texture are closely related; however, the amount of information provided by the reflectance data is limited, some other spectral features such as the absorption features of the spectral curve, such as absorption peaks, absorption valleys, etc., and the shape features of the spectral curve such as inflection point, convex point , concave points, etc., and the influence of the baseline effect on the spectral curve, etc., the reflectance data cannot be intuitively reflected; therefore, it is necessary to transform the hyperspectral reflectance data, such as normalization transformation, derivative transformation, etc.

[0004] At present, the regression methods of hyperspectral inversion soil moisture or salinity model include partial least squares method, neural network method, stepwise regression method, multiple linear regression method, etc.; among them, linear partial least squares method and nonlinear neural network method can overcome high To solve the problem of multicollinearity of spectral variables, a high-dimensional regression model is established; however, due to too many variables in the model, the efficiency of model calibration is low, and the stability of the model is poor, and the generalization is not strong; while the stepwise regression method, multivariate Linear regression methods have high requirements on the number of variables, and it is difficult to directly use high-dimensional hyperspectral data

Images