Abstract—Measurements obtained using ground penetrating radar (GPR) can suffer from large amount of noise and clutter. Current methods, such as time gating and background averaging can be applied to remove reflections from air-ground interface but do not perform well when removal of unwanted clutter signals originating from the objects other than the target is needed. This work describes and evaluates performance of two signal processing techniques – two-dimensional Principal Component Analysis (2DPCA) and Independent Component Analysis (ICA) in this kind of tasks. Experimental data using simple geometric shapes under laboratory conditions, containing strong clutter components are used to demonstrate the effectiveness of the proposed techniques.
Index Terms—Eigenimages, ground penetrating radar, independent component analysis, principal component analysis, and target signal.
The authors are with the School of Engineering, University of Portsmouth, Portsmouth, PO1 3DJ UK (e-mail: branislav.vuksanovic@port.ac.uk).
Cite: Branislav Vuksanovic, Nurul Bostanudin, Hizrin Hidzir, and Hassan Parchizadeh, "Discarding Unwanted Features from GPR Images Using 2DPCA and ICA Techniques," International Journal of Information and Electronics Engineering, vol. 3, no. 3, pp. 317-323, 2013.