GPR Target Detection Based on Fractal Dimension

Abstract:

To achieve rapid target detection by GPR (Ground Penetrating Radar),the article introduces fractal dimension as the characteristic to describe the complexity of the one-dimensional discrete signal and puts forward an algorithm for rapid target detection by GPR. First, calculate the fractal dimension of signals at different points in the direction of the survey line and depth direction after wavelet transform processing to obtain the curve of fractal dimension on two directions. Finally, determine the suspicious region in the curve of fractal dimension based on the judgment criterion to fulfill target detection. The measured data testify that the method can achieve the rapid detection on a specific target in a certain context.

• Periodical: Applied Mechanics and Materials (Volumes 321-324)
• Pages: 785-789
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.321-324.785
• Online since: June 2013
• Authors: Chang Hai Li, Yong Sheng Liang, Kong Qiang Zhu, Yan Chun Liu
• Keywords: Fractal Dimension, Ground Penetrating Radar (GPR), Target Detection

References

[1] P Gamba, S Lossani. Neural Detection of Pipe Signatures in Ground Penetrating Radar Data [J]. IEEE Trans Geoscience and Remote Sensing, 2000, 38 (2):790-797.

DOI: 10.1109/36.842008

[2] CHEN Deli, HUANG Chun-lin, SU Yi. An Integrated Method of Statistical Method and Hough Transform for GPR Targets Detection and Location [J]. ACTA ELECTRONICA SINICA, 2004, 32 (9) :1468-1471. (In Chinese)