Application of the machine learning method in an inverse scattering problem

Details Application of the machine learning method in an inverse scattering problem Application of the machine learning method in an inverse scattering problem

Dec 2002

adsabs.harvard.edu/cgi-bin/nph-data_query?bibcode=2002agufm.p72a0477s&link_type=abstract

American Geophysical Union, Fall Meeting 2002, abstract #P72A-0477

Computer Science

Numerical Analysis

0629 Inverse Scattering, 0933 Remote Sensing, 6225 Mars, 6235 Mercury

Scientific paper

In the past few years and in the next decade, several ground penetrating radars have been or will be used to explore geological composition of subsurface layers, for example, with the aim to discover watter or ice subsurface reservoirs. This paper is focused on the fundamental physical and mathematical theory behind these measurements and numerical analysis of a gradient inverse scattering method for the determination of the geometrical and dierectrical properties of the subsurface layers from received radar signal bacscattered by the studied media. We consider a model of an ellipse-shaped seam surrounded by several layers with different conductivity and a numerically generated backscattered signal (eventually with imposed artifical noise). We discuss several convergence methods (Newton's method, gradient and conjugate gradient methods, Levenberg-Marquardt method, false position method, simulated annealing and treshold accepting) used for the inverse analysis of the backscattered signal. To improve accurancy and efficiency of these methods we use machine learning method. Its procedures decision trees and support vector machines are analyzed with respect to their ability to recover seam's properties from the signal without additional information about the backscattering process.

Affiliated with

Also associated with

No associations

Rating

Application of the machine learning method in an inverse scattering problem does not yet have a rating. At this time, there are no reviews or comments for this scientific paper.

If you have personal experience with Application of the machine learning method in an inverse scattering problem, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Application of the machine learning method in an inverse scattering problem will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFWR-SCP-O-1892364