Data augmentation of ground penetrating radar images by pix2pix and its effect

Abstract

The our purpose is to develop the AI system to identify whether or not it is a risk factor for underground objects from ground penetrating radar images. So far, we have developed a method to generate a large amount of radar images by the FDTD method and then to train the AI system. However, it is difficult to generate a large amount of radar images because the FDTD method needs huge computer power. In this study, we report a method to generate a large amount of radar images similar to the FDTD method using pix2pix.