A neural network approach to the interpretation of ground penetrating radar data

Author

Costamagna, E. ; Gamba, Paolo ; Lossani, S.

Author_Institution

Dipt. di Elettronica, Pavia Univ., Italy

Volume

1

fYear

1998

fDate

6-10 Jul 1998

Firstpage

412

Abstract

A neural procedure for the analysis of ground penetrating radar images is presented. The method works by adapting the input image to the search of some objects´ signatures, that are successively identified by means of a recognition step using a back-propagation neural network. The results on actual data showing buried pipe signatures present the same degree of accuracy than the analyses performed by a human operator

Keywords

adaptive signal processing; backpropagation; geophysical signal processing; geophysical techniques; geophysics computing; neural nets; object detection; radar imaging; remote sensing by radar; terrestrial electricity; backpropagation; buried object detection; buried pipe; geoelectric method; geophysical measurement technique; ground penetrating radar; land surface; neural net; neural network; radar imaging; radar remote sensing; recognition step; terrain mapping; terrestrial electricity; Data analysis; Data mining; Ground penetrating radar; Image analysis; Neural networks; Noise reduction; Noise robustness; Performance analysis; Wavelet analysis; Wavelet domain;

fLanguage

English

Publisher

ieee

Conference_Titel

Geoscience and Remote Sensing Symposium Proceedings, 1998. IGARSS '98. 1998 IEEE International

Conference_Location

Seattle, WA

Print_ISBN

0-7803-4403-0

Type

conf

DOI

10.1109/IGARSS.1998.702923

Filename

702923