Publication:
Automatic quantitative digital analysis of concrete microstructures by 64-bit imageJ analysis software

dc.contributor.author Rider, Myca Irene en_US
dc.date.accessioned 2022-03-21T11:41:38Z
dc.date.available 2022-03-21T11:41:38Z
dc.date.issued 2010 en_US
dc.description.abstract This is the first reporting in Materials Science and Engineering at UNSW of accurate quantitative digital micro-structural analyses of 95-100mm diameter concrete cores utilising 64-bit image acquisition, and 64-bit image analysis software. The results of these newly developed, highly accurate automated quantitative analysis techniques to forensically compare micro-structural variations between hardened concrete core samples are presented. Quantitative micro-structural digital analysis of large image data files (>3 GB), have only recently been able to be studied outside the proprietary concrete laboratories with the release of 64-bit Image J, 64-bit Windows 7 operating system and 64-bit Office 2010. This thesis will demonstrate that only three simple procedures need to be followed to attain accurate quantitative digital analysis of large hardened concrete micro-structures: 1. The samples must be prepared by slow and careful fine diamond lapping with surface cooling to prevent any underlying damage to concrete micro-structure from overheating, thus preventing pull-outs of fine aggregate and sand particles within the fine cement paste matrix. This also ensures that accurately sized sharp pore perimeters remain and very fine air voids and hairline shrinkage micro-cracks are revealed. 2. The digital image needs to be obtained by a highly accurate white cold cathode fluorescent lamp IR LED light sourced digital scanner, such as a flat bed Epson Perfection V700 Photo scanner with true optical resolution of at least 6400ppi x 6400ppi (~4μ2). It will be preferable in the near future to use a much higher resolution, as for example 50800ppi x 50800ppi (0.5μ2), comparable to the best current proprietary photo scanners and saved in uncompressed TIFF file in 128-bit software, as soon as 128-bit Windows 8 becomes available. 3. All digital image files must not be altered or enhanced in any way. Any enhancement of the digital image file changes forever its ability to be used for accurate quantitative digital analysis. All image analysis software calculations thereafter can only be then roughly qualitative. Hundreds of digital image files ranging from 72ppi to 6400ppi in 32-bit and 64-bit digital analysis software were required to be able to say that the calibrated, guaranteed, and highly accurate quantitative analyses of complex microstructures from a variety of hardened concrete mixes were correct. The 64-bit Image Analysis and Processing Software enabled me to develop automated, highly accurate, forensic comparative methods to study the macro-structures and micro-structures of commercial sized concrete core samples. en_US
dc.identifier.uri http://hdl.handle.net/1959.4/52211
dc.language English
dc.language.iso EN en_US
dc.publisher UNSW, Sydney en_US
dc.rights CC BY-NC-ND 3.0 en_US
dc.rights.uri https://creativecommons.org/licenses/by-nc-nd/3.0/au/ en_US
dc.subject.other quantitative en_US
dc.subject.other digital analysis en_US
dc.subject.other automatic en_US
dc.subject.other concrete en_US
dc.subject.other microstructures en_US
dc.subject.other quantatitive en_US
dc.subject.other 64-bit image J en_US
dc.title Automatic quantitative digital analysis of concrete microstructures by 64-bit imageJ analysis software en_US
dc.type Thesis en_US
dcterms.accessRights open access
dcterms.rightsHolder Rider, Myca Irene
dspace.entity.type Publication en_US
unsw.accessRights.uri https://purl.org/coar/access_right/c_abf2
unsw.identifier.doi https://doi.org/10.26190/unsworks/15770
unsw.relation.faculty Science
unsw.relation.originalPublicationAffiliation Rider , Myca Irene, Materials Science & Engineering, Faculty of Science, UNSW en_US
unsw.relation.school School of Materials Science & Engineering *
unsw.thesis.degreetype Masters Thesis en_US
Files
Original bundle
Now showing 1 - 1 of 1
No Thumbnail Available
Name:
whole.pdf
Size:
22.98 MB
Format:
application/pdf
Description:
Resource type