Computer-aided versus manual grain size assessment in a single phase material

The results of grain size assessment obtained using computer image analysis and manually are presented for comparison. Twelve images of austenitic ste el structures of different grain size were examined. The most important gra in size parameters, i.e., mean cross section and mean grain diameter were d etermined. To verify the proposed algorithms the results obtained for the s ame structures by using classical quantitative metallography are also prese nted. The algorithms based on significantly different methods for grain edg e detection were used to demonstrate the relationships more clearly and to provide a more general view of computer-aided image processing. The measure ments were also made manually by six independent observers. The comparison between different methods indicates an advantage of automatic techniques. ( C) 2001 Elsevier Science Inc. All rights reserved.