Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Effects of different pretreatment and modeling methods on soil moisture content detected by near infrared spectroscopy

Published by the American Society of Agricultural and Biological Engineers, St. Joseph, Michigan

Citation:  2021 ASABE Annual International Virtual Meeting  2100335.(doi:10.13031/aim.202100335)
Authors:   Xin Gao, Bing Lv, Ke He, Xiuying Tang
Keywords:   partial least squares; pretreatment method;Near infrared; soil water content


Soil moisture is the main source of water needed by plants, an important factor affecting agricultural operation measures and precision agriculture, and a necessary parameter in the research process of agriculture, environment and other disciplines. Therefore, accurate, simple and rapid measurement of soil water content is very important. In this paper, near infrared spectroscopy was used to realize the rapid detection of soil moisture. A total of 120 soil samples with different water contents were prepared, of which 113 were available samples. According to the reflectance spectral data of samples in 900-1700 nm band, the spectral data were preprocessed by different methods, including first derivative, second derivative, Savitzky-Golay ( S-G ) smoothing, Savitzky-Golay smoothing combined with first derivative, Savitzky-Golay smoothing combined with second derivative, standard normal variable transformation ( SNV ), multiplicative scatter correction ( MSC ), and then the partial least squares ( PLSR ) method was used to establish the relationship model between the spectral data and the measured water content. The results show that the partial least squares model with the best prediction effect on soil moisture content can be established by using S-G smoothing method to preprocess the spectral data.

(Download PDF)    (Export to EndNotes)