Title :
Inversion of the soil moisture based upon neural network
Author :
Zongqian, Li ; Yuhua, Tu ; Ning, LIU
Author_Institution :
Dept. of Electron. Eng., Tsinghua Univ., Beijing, China
Abstract :
An inversion method based upon a neural network is applied to retrieve the soil moisture and roughness parameters from the radar backscattering coefficients. The structure and the training algorithm of the neural network (NN) are presented in the paper where the training patterns are generated by an integral equation model (IEM). By choosing the proper type of input data, the necessary input data number is minimized. Analyze of the calculation results shows that the NN inversion method has high accuracy.
Keywords :
backscatter; geophysical signal processing; hydrological techniques; integral equations; inverse problems; moisture; neural nets; radar signal processing; remote sensing by radar; rough surfaces; soil; input data; input data number; integral equation model; inversion method; neural network; radar backscattering coefficients; roughness parameters; soil moisture; training algorithm; Backscatter; Geometry; Integral equations; Multi-layer neural network; Neural networks; Radar scattering; Scanning probe microscopy; Soil measurements; Soil moisture; Spaceborne radar;
Conference_Titel :
Antennas, Propagation and EM Theory, 2000. Proceedings. ISAPE 2000. 5th International Symposium on
Conference_Location :
Beijing, China
Print_ISBN :
0-7803-6377-9
DOI :
10.1109/ISAPE.2000.894808