
Evaluating Mechanical Properties of Carbonate Rocks Using Hyperspectral Imaging in the NIR Region
Anton Vernytsky (1), Dr. Dagan Bakun Mazor (2), Dr. Isaac August (3)
(1) Other Institute (insert manually)
(2) Department of Civil Engineering, SCE - Shamoon College of Engineering, Beer-Sheva 8410802, Israel.
(3) Other Institute (insert manually)
(1) Department of Green Engineering, SCE – Shamoon College of Engineering, Beer-Sheva 8410802, Israel. (3) Department of Electrical Engineering, SCE – Shamoon College of Engineering, Beer-Sheva 8410802, Israel.
Carbonate rocks are the primary source of natural aggregates for concrete and infrastructure in Israel, and their performance depends on key physical and mechanical properties, including uniaxial compressive strength (UCS), density, porosity, and water absorption. These parameters are commonly obtained through destructive laboratory testing, which is time-consuming, costly, and limited in representativeness. This study evaluates hyperspectral imaging (HIS) as a rapid, non-destructive approach for estimating these properties from spectral information linked to carbonate mineralogy and microstructure. A total of 168 carbonate rock samples spanning a wide range of strengths were collected from various outcrops across Israel and tested in the laboratory for UCS, density, porosity, and water absorption. Hyperspectral data were acquired in the near-infrared range (900-1700 nm) using a hyperspectral camera (SPECIM FX17) under both controlled laboratory conditions and outdoor natural illumination. Spectral data were processed to minimize illumination effects and enhance diagnostic spectral features. Then predictive models were developed to relate hyperspectral information to measured rock properties. The results demonstrate strong predictive capabilities for density, porosity, and water absorption, and meaningful, though weaker, estimation of UCS. Differences between laboratory and outdoor measurements reflect the influence of illumination and atmospheric effects, highlighting challenges and opportunities for field-based applications. The findings indicate that hyperspectral imaging provides a promising tool for rapid, non-destructive assessment of carbonate rocks properties and may support applications in quarrying, engineering geology, and mineral resource evaluation in Israel.



