Abstract
Vickers hardness is still measured by human operators for accurate measurement, because automatic measurement sometimes shows poor accuracy due to the slight difference in contrast and shape of the indentation. In this study, for more accurate Vickers hardness automatic measurement, we propose a novel technique by using convolutional neural network (CNN). We examine the usefulness of our novel technique, compared with manual measurement and image processing measurement. The hardness values measured by the CNN method suggest being close to the values measured manually, and more accurate than the image processing method.
Export citation and abstract BibTeX RIS
Content from this work may be used under the terms of the Creative Commons Attribution 3.0 licence. Any further distribution of this work must maintain attribution to the author(s) and the title of the work, journal citation and DOI.