Underground Anomaly Detection in GPR Data by Learning in the C3 Model Space

Ground-penetrating radar (GPR) provides an effective means for underground anomaly detection, but it is also accompanied by some practical issues such as the lack of prior knowledge, insufficient labeled data, and timeliness constraints. In this article, we propose detecting underground anomalies by adequately capturing multidirectional changing information within GPR B-scan data, which extends the framework of model-space learning (MSL). MSL aims to transform the data from the data space into the model space by representing the original data with fit models that capture the changes within the data. In GPR data, due to the continuity of the underground medium and electromagnetic wave, there is effective changing information in each column of data along not only the vertical direction but also the horizontal detecting direction according to the subsurface environments and existing underground structures. To fully capture the changing information within GPR data, we fit the GPR data along multiple directions, respectively, and synthesize the fit models into a comprehensive-change-captured model (C3 model) wherein multidirectional changing information within the original data is captured. Representing the original data with the C3 model transforms the data from the data space into the C3 model space. The distance metric between C3 models is then introduced, and learning methods could be efficiently implemented on the models. Experiments are conducted on GPR data collected along urban roads, with or without prior knowledge about the existing subsurface anomalies. The obtained results demonstrate the effectiveness of the proposed method.