Advanced Computational Approaches for GPR Processing and Interpretation

Ground penetrating radar is a well-established technique for civil engineering applications. However, the method still needs improvements and new developments when interpreting the reflection profiles produced. Advances are therefore required both for automated detection and better understanding of the GPR signal propagation. The development of new algorithms to automate the process of interpretation and detection of possible targets allows minimizing efforts, which reduces the amount of time needed to take a decision. In addition, the interpretation of the GPR results can become complicated when diffraction events or reflection multiples occur. In such circumstances, the use of sophisticated modeling of the electromagnetic signal is recommended to assist in the interpretation of field data. Thus, this work presents the development of new algorithms for automated detection and processing when measuring pavement thicknesses, as well as new numerical modeling approaches to improve the interpretation process when surveying complex and heterogeneous structures.