Automated grain boundary detection and classification in orientation contrast images

An unbiased and unequivocally defined estimate of grain sizes and shapes is fundamental for understanding the microscopic behaviour of crystalline materials modified by the action of stress fields and/or chemical gradients. Because of their very good spatial resolution, orientation contrast (OC) images represent a useful starting point to develop an automated technique able to assess grain boundaries in a completely objective and reproducible way. The method presented in this contribution defines boundaries as high brightness gradient features on an OC image of a quartz mylonite through a specifically designed sequence of detection and filter algorithms that minimise the effect of local background noise. The object set into which the OC image has been divided is further analysed to compute a set of positions where to perform electron backscatering diffraction analysis and build a crystal orientation data set. This data set is then used along with information from the detection-filtering algorithm to automatically rebuild the real grain boundary net. The obtained results are in good agreement with results from similar manual techniques, while the whole determination process is also much faster than other automated electron backscattering diffraction analytical methods.