Adaptive Subsurface Visualization System Using Phase Retrieval Method and Complex-Valued Self-Organizing Map

Abstract

We propose an adaptive subsurface visualization system based on a complex-valued self-organizing map (CSOM). Conventionally buried things can be detected in so-called B-scan images obtained by a ground penetrating radar. In contrast, our proposed method is able not only to detect their presence, but also to classify the targets by the self-organizing dynamics in the CSOM. Instead of utilizing only the amplitude information in the time domain, we use both the amplitude and phase information to obtain the scattering coefficients of the targets by use of the phase retrieval method. The CSOM classifies the obtained scattering coefficients to realize an automatic categorization of the targets that scatter the radar electromagnetic wave.