A novel processing framework for tree root mapping and density estimation using ground penetrating radar

Estimating the root density of mature trees is of high importance as the root system is a robust indicator of both the health status and the structural integrity of a tree. With this in mind, a multi-stage data processing scheme is proposed using Ground Penetrating Radar (GPR) to achieve an effective estimation of the root density of trees. The proposed framework is divided into three main chronological steps. Initially, ringing noise is removed using a Singular Value Decomposition (SVD) filter prior to a frequency-wavenumber (F-K) migration. Subsequently, a tracking algorithm is applied to the processed data in an effort to identify patterns associated with roots. Lastly, the found patterns are expressed as continuous and differentiable functions from which the root density is derived. To demonstrate the viability of the proposed approach, a case study is presented in order to identify the root system and map the overall density of the roots of a mature tree. The algorithm is commercially appealing with minimum computational and operational requirements for large-scale forestry applications.