Image processing and analysis of the grain constitution in minerals

Abstract Methods are presented for the postprocessing and analysis of microscopic and macroscopic images to investigate mineral grain constitution. An algorithm for eliminating indicated objects is described. It is used both in procedures to eliminate marginal and small objects and to fill gaps, and in the organization of grey-level and binary image analysis. Image processing algorithms which work without operator control have been implemented in simple and low-cost 16-bit uniprocessor image systems.