Implementation of time domain filtering in the complex wavelet framework for GPR noise attenuation

GPR is one of the geophysical methods that uses high frequency non-stationary electromagnetic waves to image underground structures. The transported GPR data is generally polluted by various types of noise caused by other equipment emitting radio waves or man-made structures near the GPR system. Thus, improving the capability of causal temporal filtering as simple and efficient algorithms for removing noise from GPR signals is proposed in this paper. In this study, time-frequency analysis methods including Savitzky-Golay (SG) time domain filter and Dual-Tree Complex Wavelet Transform (DTCWT) to attenuate noise from GPR data has been investigated. Non-Gaussian noises were targeted for attenuation and ultimately led to the presentation of a new method of noise attenuation and signal preservation in sections obtained from the GPR method by designing a filter in a complex two-dimensional wavelet frame.