Using Random Forest Tree Classification for Evaluating Vertical Cross-Sections in Epoxy Blocks to Get Unbiased Estimates for 3D Mineral Map

Areal mineral maps are constructed from the polished sections of particles that settle to bottom epoxy resin. However, heavy minerals can preferentially bottom, making surface rich in minerals. Therefore, will become biased estimates volumetric (3D) map. The study aims test whether any vertical cross-section (any section along settling direction particles) be an unbiased estimate 3D map a chromite ore sample. For purpose this study, 2D cross-sections were acquired by using Random Forest classification coupled with image pre- and post-processing tools. Then, converted without assuming stereological errors. modal mineralogy particle size distributions predicted compared same features estimated particulate sample XRD dry sieving analyses, respectively. Any which yields distribution similar true analyses was selected as results suggest is estimate, unlike at heavier preferentially.