A method for the digital image analysis of ceramic grains based on shape factor segmentation

Abstract

A very simple and robust method for ceramics grains quantitative image analysis is presented. Based on the use of optimal imaging conditions for reflective light microscopy of bulk samples, a digital image processing routine was developed for shading correction, noise suppressing and contours enhancement. Image analysis was done for grains selected according to their concavities, evaluated by perimeter ratio shape factor, to avoid consider the effects of breakouts and ghost boundaries due to ceramographic preparation limitations. As an example, the method was applied for two ceramics, to compare grain size and morphology distributions. In this case, most of artefacts introduced by ceramographic preparation could be discarded due to the use of perimeter ratio exclusion range.