Sparse signal decomposition for ground penetrating radar

Authors

Wenbin Shao, University of WollongongFollow
Abdesselam Bouzerdoum, University of WollongongFollow
Son Lam Phung, University of WollongongFollow

RIS ID

37293

Publication Details

W. Shao, et al., "Sparse signal decomposition for ground penetrating radar," in 2011 IEEE Radar Conference, Kansas City, Missouri, 2011, pp. 453 - 457.

Abstract

In this paper, we present an adaptive approach for sparse signal decomposition, in which each GPR trace is decomposed into elementary waves automatically. A sparse feature vector is extracted from the decomposition and used for classification of railway ballast. The experimental results show that the proposed approach can represent the GPR signals efficiently, and effective features can be extracted for pattern classification.

ANZSRC / FoR Code

080109 Pattern Recognition and Data Mining, 090609 Signal Processing