A Bayesian approach to retrieve soil parameters from SAR data: Effect of prior information

Soil moisture retrieval from SAR images is always affected by speckle noise, model errors and uncertainties associated to soil parameters, which impact negatively on the accuracy of soil moisture estimates. A Bayesian approach has been proposed to deal with these issues. As a natural advantage of th...

Descripción completa

Guardado en:
Detalles Bibliográficos
Autores principales: Barber, M., Maas, M., Perna, P., Grings, F., Karszenbaum, H.
Formato: CONF
Materias:
Bayesian methods
Inverse problems
Prior information
Radar applications
Soil moisture
Synthetic aperture radar
Bayesian approaches
Soil conditions
Soil moisture retrievals
Soil parameters
Speckle noise
Bayesian networks
Computer simulation
Geologic models
Image analysis
Uncertainty analysis
Acceso en línea:http://hdl.handle.net/20.500.12110/paper_0277786X_v8536_n_p_Barber
Aporte de:
Biblioteca Digital - Facultad de Ciencias Exactas y Naturales (UBA) de Universidad de Buenos Aires
Descripción
Sumario:Soil moisture retrieval from SAR images is always affected by speckle noise, model errors and uncertainties associated to soil parameters, which impact negatively on the accuracy of soil moisture estimates. A Bayesian approach has been proposed to deal with these issues. As a natural advantage of the Bayesian approach, prior information about soil condition can be easily included. Based on simulations, the effect of prior information has been analyzed. It follows from simulations using the Oh's model that the soil moisture estimator is very sensitivity to the roughness prior. © 2012 SPIE.

Ejemplares similares