Application of electrical capacitance tomography – spectroscopy to predict the internal conditions of tree trunks

Abstract

Non-destructive monitoring of the internal condition of tree trunks has been widely implemented. However, existing methods often cause minor injuries to the trunks, which is undesirable for the wood industry in plantation forests. This study aims to develop a damage-free scanning technology to assess the internal condition of tree trunks, utilizing a combination of electrical capacitance tomography and spectroscopy. The approach was conducted by scanning felled timber at sawmills. The selected samples were in intact condition as well as those with defects such as cavity and decay. Data collection was performed using a prototype of capacitance-based tomography equipped with a belt-shaped sensor. Dielectric data were obtained from segments every 30 cm along the trunk's length and subsequently processed using machine learning. The analysis indicated significant differences in electrical characteristics between intact and defective wood. Intact wood displayed the highest dielectric values, while defective wood exhibited varying dielectric values depending on its condition, with the lowest values associated with 26% cavity. These results suggest that intact wood tissue has favorable permittivity properties, leading to higher dielectric values. This study demonstrates the potential of the developed technology to predict the internal condition of tree trunks through the analysis of dielectric values.