Real-time imaging system for a ground penetrating radar.

摘要

This thesis introduces an imaging system that generates two-dimensional radar images in real time for the detection of shallow buried objects in homogeneous sand. The main imaging problem is that the plastic dielectric constant is close to that from the sand, which causes a small contrast between their reflections. Furthermore, the reflected signals contain information from the ground surface and other buried objects that obscure the reflections from the targeted object. Therefore, in order to contribute to the development of practical systems for detecting buried non-metallic objects, a signal processing algorithm was developed in this work. The algorithm generates complex range profiles by applying the Inverse Fourier Transform (IFFT) to the data and corrects the position shift of the scatterers due to the GPR system delay. Afterwards, the first difference is applied to the data in order to enhance the final image and discriminate the response transitions between the targeted object and the sand. (Abstract shortened by UMI.)