Logo image
Back
Heartwood detection in sandalwood (Santalum album) using resistance drilling
Preprint   Open access

Heartwood detection in sandalwood (Santalum album) using resistance drilling

Duc Bui, David J Bush, Geoffrey M Downes, Vilius Gendvilas and David Lee
Research Square, Vol.5 August 2026
Research Square Company
2026
DOI: https://doi.org/10.21203/rs.3.rs-10531485/v1
pdf
v1_covered_ed5cfbfb-ced3-4d99-a0c6-4bb5e866083d2.11 MBDownloadView
Preprint Version Open Access CC BY V4.0
url
https://www.researchsquare.com/article/rs-10531485/v1View
Preprint Version CC BY V4.0

Abstract

Santalum album heartwood detection resistance drilling feature-based approach
Heartwood is the most valuable component of sandalwood species (Santalum spp.), yet its detection in standing trees remains challenging due to reliance on destructive detection methods. This study evaluates the resistance drilling (RESI) method for detecting heartwood in Santalum album trees. RESI provides continuous one-dimensional radial resistance profiles including drilling resistance (DR) and forward resistance (FR) across drilling depth. This is the first study to summarize DR and FR profiles by “features”, including Mean, Top 10%, Coefficient of Variation (CV), Skewness, Rugosity, and the ratio of mean DR/FR. This approach transformed RESI profiles into vectors describing the spatial structure of internal wood, enabling classification of heartwood relative to sapwood at both between-tree and within-tree. Among trees, Mean of DR was the primary detecting feature with those containing heartwood showing higher mean of DR than those without. Within trees, CV, Skewness, and Rugosity were RESI features that significantly differed between sapwood and heartwood. Among them, CV of DR was the best detecting feature, where heartwood had lower heterogeneity than sapwood. Using feature-based approach, RESI provides a semi-automated method for assessing sandalwood heartwood, leveraging its high sampling rate and minimally invasive measurements. However, classification thresholds varied across tree ages and planting systems. Larger datasets that capture this variability would enable the establishment of robust baselines.

Details

Metrics

1 Record Views
Logo image