Demonstration of Optical Microscopy and Image Processing to Classify Respirable Coal Mine Dust Particles

Respirable coal mine dust represents a serious health hazard for miners. Monitoring methods are needed that enable fractionation of into its primary components, and do so in real time. Near the production face, simple capability to monitor versus mineral fractions would be highly valuable tracking changes sources—and supporting timely responses terms controls or other interventions reduce exposures. In this work, premise monitoring with polarized light microscopy was explored. Using images representative particles (kaolinite, crystalline silica, limestone rock dust), model built exploit birefringence effectively separate them from coal. The showed >95% accuracy on test dataset known particles. For composite samples containing both minerals, also very good agreement results scanning electron classification, which used as reference method. Results could further concept “cell phone microscope” type semi-continuous measurements mines.