GrainAutLine: an Environment for Semi-Automatic Processing of Marble Thin Section Images

  • Kristóf Csorba Department of Automation and Applied Informatics, Faculty of Electrical Engineering and Informatics, Budapest University of Technology and Economics, Hungary
  • Ádám Budai Department of Automation and Applied Informatics, Faculty of Electrical Engineering and Informatics, Budapest University of Technology and Economics, Hungary
  • Judit Zöldföldi Institute of Material Science, University of Stuttgart, Stuttgart, Germany
  • Balázs Székely Department of Geophysics and Space Sciences, Eötvös Loránd University, Budapest, Hungary

Abstract

GrainAutLine is an interdisciplinary microscopy image analysis tool with domain specific smart functions to partially automate the processing of marble thin section images. It allows the user to create a clean grain boundary image which is a starting point of several archaeometric and geologic analyses. The semi-automatic tools minimize the need for carefully drawing the grain boundaries manually, even in cases where twin crystals prohibit the use of classic edge detection based boundary detection. Due to the semi-automatic approach, the user has full control over the process and can modify the automatic results before finalizing a specific step. This approach guarantees high quality results both in cases where the process is easy to automate, and also if it needs more help from the user. This paper presents the basic operation of the system and details about the provided tools as a case study for an interdisciplinary, semi-automatic image processing application.

Keywords

marble, thin section, grain boundary, semi-automatic segmentation, software tool
Published in Onlinefirst
17-08-2017
How to Cite
CSORBA, Kristóf et al. GrainAutLine: an Environment for Semi-Automatic Processing of Marble Thin Section Images. Periodica Polytechnica Electrical Engineering and Computer Science, [S.l.], v. 61, n. 4, p. 305-311, 2017. ISSN 2064-5279. Available at: <https://pp.bme.hu/eecs/article/view/10890>. Date accessed: 21 nov. 2017. doi: https://doi.org/10.3311/PPee.10890.
Section
Articles