Application of neural network and principal component analysis to GPR data

Abstract

This communication presents a prediction algorithm for the detection of features in GPR-based surveys. Based on signal processing and soft-computing techniques, the coupled use of principal-component analysis and neural networks enables a definition of an efficient method for analyzing GPR electromagnetic data. Results for detecting features of geological layers demonstrate not only the accuracy of the predictions of the method but also the simple interpretation of its output through reconstructed images of the scenarios.