3. Conclusion
In construction of Well-facilitied capital farmland, in addition to
emphasize on land leveling, roads, canals and other engineering
facilities, and soil quality after completion reaches the higher level
of the county, and the soil PH, SOM, soil nutrient, and soil pollution
characteristic is indispensable basic information of cultivated land
quality evaluation, it is also an important guarantee for the growing
environment of crops. Therefore, the fast acquisition and real-time
monitoring of soil basic information is the premise and foundation for
the construction of Well-facilitied capital farmland.
Based on this, this research took the Well-facilitied capital farmland
construction area in Xinzheng City as the research object, combined
field sampling and indoor hyperspectral determination, conducted
correlation analysis and Fuzzy clustering analysis of soil properties
and SG-CR transform spectrum, and selected the best common
hyperspectral characteristic band. And constructed the panel data model
of hyperspectral comprehensive inversion about 116 samples of 11 soil
attribute, to achieve comprehensive and rapid acquisition of soil basic
information. The results show that:
(1) SG convolution smoothing of spectral reflectance can
effectively remove noise, while preserving the overall characteristics
of spectral curves, and improving signal-to-noise ratio. The method of
spectral transformation is very important, for extracting characteristic
bands and improving the prediction ability and stability of the model.
By removing the envelope CR spectral transformation, the
absorption and reflection characteristics of the spectral curve are
effectively highlighted, the sensitivity of the spectrum is improved,
and the useful information of the spectrum is enhanced.
(2)After the correlation analysis about soil attributes and SG-CRspectral transformation, and using Fuzzy clustering maximum tree method,
combined with the similar inflection point of the correlation
coefficient curve of soil attributes and spectral reflectance, select
the common significance band of different soil attributes as the best
hyperspectral characteristic band and focus on
405~431nm、781nm~831nm、1044~1087nm、1251~1410nm、1836~1898nm、2080nm~2201nm、2324~2395nm.
(3) Based on the panel data model, constructed the comprehensive
inversion model of soil properties in the well-facilitied capital
farmland construction area of Xinzheng city, with the common spectral
characteristic band of SG-CR spectral transformation as the
independent variable. The model was significant as a whole, and the
goodness of fit was high(2 = 0.9991,DW =
2.1899,F = 2195.67). The coefficientR̅ 2 of determination of 11 soil attributes were
all greater than 0.95, and the root mean square root was small. From the
perspective of relative analysis errors, the relative analysis errors of
soil attributes of SG-CR spectral transformation model are all
greater than 2.5, and the measured and predicted values of most samples
are concentrated near the 1:1 line, and the correlation coefficientsr all pass the significance test at P = 0.01. It indicates
that the panel data models with SG-CR spectral transformation as
independent variable have the ability of comprehensive inversion of soil
properties, and have high precision prediction.