Publication:
A cascade hyperbolic recognition of buried objects using hybrid feature extraction in ground penetrating radar images
A cascade hyperbolic recognition of buried objects using hybrid feature extraction in ground penetrating radar images
Date
2021-08-27
Authors
Hasimah Ali
Ahmad Firdaus Ahmad Zaidi
Wan Khairunizam Wan Ahmad
Mohd Shuhanaz Zanar Azalan
Tengku Amran T.S.
Ahmad M.R.
Mohamed Elshaikh Elobaid Said Ahmed
Journal Title
Journal ISSN
Volume Title
Publisher
Research Projects
Organizational Units
Journal Issue
Abstract
Ground penetrating radar (GPR) has been acknowledged as effective nondestructive technique for imaging the subsurface. But the process of recognizing hyperbolic pattern of buried objects is subjective and mainly relies upon operator's knowledge and experience. This project proposed a hyperbolic recognition of buried objects using hybrid feature extraction in GPR subsurface mapping. In this framework, a cascade hyperbolic recognition by means of Empirical Mode Decomposition (EMD) and Discrete Wavelet Transform (DWT) are used as hybrid feature recognizing hyperbolic of buried objects. The rationale for an initial focus on cascade hyperbolic recognition is motivated by unique features exhibits by EMD and DWT behaviour in characterizing the hyperbolic pattern which make them particularly well suited to utilities detection in GPR. A series of experiments has been conducted on hyperbolic pattern based on hybrid features using four different geometrical shapes of cubic, cylindrical disc and spherical. Based on the results obtained, the hybrid features of IMF1+ wavelet transform (cH1) shows promising recognition rate in recognizing the hyperbolic that having different geometrical shapes of buried objects.