Signal processing for tree-trunk investigation using ground penetrating radar

Invasive fungi diseases are considered one the biggest threats for the ash and oak forests in United Kingdom. To that extend, Ground Penetrating Radar (GPR) can provide a powerful diagnostic tool for assessing the health status of tree trunks based on their internal dielectric distribution. GPR acquisitions in tree-trunks is a unique problem that can not be approached with traditional GPR processing approaches. Typical interpretation tools like hyperbola fitting and migration should be adjusted and fine-tuned in order to be applicable for irregular measurements in a closed curve. The purpose of this paper is to provide GPR practitioners with a set of interpretation tools that can be applied in the field using commercial GPR antennas. In that context, a novel processing framework is presented that is fine-tuned for the current problem. The suggested scheme is successfully tested using both numerical and real data indicating the capabilities of GPR as a diagnostic tool for early detection of tree diseases.