Image Processing of Porous Structures in Iron Ore Pellets
Independent thesis Advanced level (professional degree), 20 credits / 30 HE creditsStudent thesis
In this thesis, image processing tools are used for analysing properties of iron ore pellets, with the goal of increased understanding of the balling process used in pellet production. Two different types of data are examined to study different aspects of the porosity in pellets.The first type of data, scanning electron microscopy images of pellet cross sections, is used for anaylsing tunnel porosity structures. An improvement to an existing analysis method for calculating the porosity network tunnel widths has been implemented.The improvements presented in this thesis does not significantly alter the end results of the analysis, since the differences in measured tunnel widths are only improved by a relatively small fraction. The actual distribution of the measured tunnel widths is not affected. The methods were tested in various way to highlight the performance in specific situations.The second type of data is 3D X-ray microtomography images of pellets. The images are analysed with the intent of labelling regions of the pellet as either a part of the pellet core or the pellet shell. Multiple convolutions, in a spherical coordinate representation of the pellet, are used to filter out and clarify regions that resemble a layered structure with locally varying porosity.The layered structures are to a greater extent present in the shell compared to the core.The presented method can not, by itself, make a clear distinction between the core and the shell – instead it turned out to be a tool more suitable for visually examining layered structures in pellets. Extensive work was done to test and validate the performance of the method.
Place, publisher, year, edition, pages
2012. , 72 p.
Technology, Porosity, iron ore pellets, image processing, moving average, 3D x-ray microtomography, porous structures, layers, balling
IdentifiersURN: urn:nbn:se:ltu:diva-51428Local ID: 8a1d0bf0-7328-43e8-9b7f-87fa737d7fefOAI: oai:DiVA.org:ltu-51428DiVA: diva2:1024789
Subject / course
Student thesis, at least 30 credits
Computer Science and Engineering, master's level
Validerat; 20120305 (anonymous)2016-10-042016-10-04Bibliographically approved