Techniques and applications for predictive metallurgy and ore characterization using optical image analysis

Abstract

Optical image analysis (OIA) as a predictive metallurgical tool has been advanced by work completed at SGS Mineral Services by integrating a standard image analysis system with innovative preparation methods and measurement techniques. The preparation of sized material into non-touching particle polished sections as well as the selective coating of epoxy facilitates the recognition of non-opaque minerals, thus correctly identifying locked and complex particles. Advanced data analysis enables the preparation of predictive metallurgical data. Test cases were compared to metallurgical test results including size distribution, mineral release and grade/recovery data for various ore deposits. The limiting factor in direct optical image analysis is the discrimination between minerals with similar reflective properties. Advances made at the University of Liege have made it possible to optically distinguish these minerals using multispectral imaging. Integrating these technologies will enable OIA to be an affordable and viable alternative for gathering process mineralogical data.