Home // International Journal On Advances in Software, volume 10, numbers 3 and 4, 2017 // View article

Compositing Ground Penetrating Radar Scans of Differing Frequencies for Better Depth Perception

Authors:
Roger Tilley
Hamid Sadjadpour
Farid Dowla

Keywords: Ground Penetrating Radar; Expectation Maximization; Gaussian Mixture Model; Maximum Likelihood parameter estimation; Finite Difference Time Domain Method, GprMax.

Abstract:
Methods developed to reduce interference in a noisy environment, be it radar target responses or effective communications in the presence of noise for mobile phone users, are vital in delivering a clear usable signal. The methods used to render a cleaner signal can also be used to combine signals of various frequencies. Ground Penetrating Radar (GPR) scans over the same area are no exception. This paper explores using an optimization problem solver, the Expectation Maximization (EM) Algorithm, to define the weights to use to combine multiple GPR scans at different frequencies over the same target area. This approach exploits the Gaussian Mixture Model (GMM) feature of the EM algorithm to produce a cleaner image at depth. Our method demonstrates a measured improvement toward producing a cleaner image.

Pages: 413 to 431

Copyright: Copyright (c) to authors, 2017. Used with permission.

Publication date: December 31, 2017

Published in: journal

ISSN: 1942-2628