The prediction method of soil moisture content based on multiple regression and RBF neural network

In the application of the field repair in the countryside, determination of the moisture content is very important. Compared with the traditional methods, ground penetrating radar (GPR) can measure soil moisture content in a wide region at the same time. This paper presents a prediction method about the moisture content based on the multiple regression and the radial basis function (RBF) neural network. Firstly, we measured the moisture content by experiments and compared the information in GPR data. Secondly, we use multiple regression analysis to get the active components affecting the moisture content in GPR data. Through utilizing the active components and the soil moisture content, we can train RBF neural network. Finally, optimize and record the network. In the practical application, aiming at a particular frequency of GPR, through multiple regression, we can predict the soil moisture content better than the RBF neural network only. This method not only can meet the needs of determination of the soil moisture content, but also can make a necessary help for the field repair in the countryside.

[1]  Bernhard Schölkopf,et al.  Comparing support vector machines with Gaussian kernels to radial basis function classifiers , 1997, IEEE Trans. Signal Process..

[2]  Lorenzo Bruzzone,et al.  A technique for the selection of kernel-function parameters in RBF neural networks for classification of remote-sensing images , 1999, IEEE Trans. Geosci. Remote. Sens..

[3]  Guirong Liu,et al.  A point interpolation meshless method based on radial basis functions , 2002 .

[4]  J. D. Redman,et al.  An analysis of the ground‐penetrating radar direct ground wave method for soil water content measurement , 2003 .

[5]  Sevket Demirci,et al.  A synthetic aperture radar‐based focusing algorithm for B‐scan ground penetrating radar imagery , 2007 .

[6]  Hojjat Adeli,et al.  Principal Component Analysis-Enhanced Cosine Radial Basis Function Neural Network for Robust Epilepsy and Seizure Detection , 2008, IEEE Transactions on Biomedical Engineering.

[7]  Xiaoyun Zhang,et al.  Microwave Imaging of Soil Water Diffusion Using the Linear Sampling Method , 2011, IEEE Geoscience and Remote Sensing Letters.

[8]  Eslam Pourbasheer,et al.  QSAR study of Nav1.7 antagonists by multiple linear regression method based on genetic algorithm (GA–MLR) , 2013, Medicinal Chemistry Research.

[9]  David Hodgetts,et al.  Laser scanning and digital outcrop geology in the petroleum industry: A review , 2013 .

[10]  Abdesselam Bouzerdoum,et al.  Sparse Representation of GPR Traces With Application to Signal Classification , 2013, IEEE Transactions on Geoscience and Remote Sensing.

[11]  Xu Qiao,et al.  The Research of the Ground Penetrating Radar Wave Velocity Estimation Method Based on QR Decomposition , 2014 .

[12]  Alex Alexandridis,et al.  Large Earthquake Occurrence Estimation Based on Radial Basis Function Neural Networks , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[13]  Enes Yigit,et al.  A Review on Migration Methods in B-Scan Ground Penetrating Radar Imaging , 2014 .

[14]  Carlos López-Martínez,et al.  Atmospheric Phase Screen Compensation in Ground-Based SAR With a Multiple-Regression Model Over Mountainous Regions , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[15]  Sébastien Lambot,et al.  Full-Wave Modeling of Near-Field Radar Data for Planar Layered Media Reconstruction , 2014, IEEE Transactions on Geoscience and Remote Sensing.

[16]  Angelika Humbert,et al.  Complex network of channels beneath an Antarctic ice shelf , 2014 .