%0 Conference Proceedings
%4 sid.inpe.br/sibgrapi/2018/10.19.14.02
%2 sid.inpe.br/sibgrapi/2018/10.19.14.02.11
%T Evaluation of segmentation methods based on classification patterns for micro-tomography applications in rock analysis
%D 2018
%A Vasconcelos, Gustavo José Querino de,
%A Antonieti, Giovanna,
%A Libel, Gustavo,
%A Rosa, Paola,
%A Archilha, Nathaly,
%A Carvalho, Tiago,
%A Pedrini, Helio,
%A Spina, Thiago Vallin,
%@affiliation Institute of Computing - UNICAMP
%@affiliation Brazilian Synchrotron Light Laboratory
%@affiliation Federal University of Technology - UTFPR
%@affiliation Brazilian Synchrotron Light Laboratory
%@affiliation Brazilian Synchrotron Light Laboratory
%@affiliation IFSP
%@affiliation Institute of Computing - UNICAMP
%@affiliation Brazilian Synchrotron Light Laboratory
%E Ross, Arun,
%E Gastal, Eduardo S. L.,
%E Jorge, Joaquim A.,
%E Queiroz, Ricardo L. de,
%E Minetto, Rodrigo,
%E Sarkar, Sudeep,
%E Papa, João Paulo,
%E Oliveira, Manuel M.,
%E Arbeláez, Pablo,
%E Mery, Domingo,
%E Oliveira, Maria Cristina Ferreira de,
%E Spina, Thiago Vallin,
%E Mendes, Caroline Mazetto,
%E Costa, Henrique Sérgio Gutierrez,
%E Mejail, Marta Estela,
%E Geus, Klaus de,
%E Scheer, Sergio,
%B Conference on Graphics, Patterns and Images, 31 (SIBGRAPI)
%C Foz do Iguaçu, PR, Brazil
%8 29 Oct.-1 Nov. 2018
%I Sociedade Brasileira de Computação
%J Porto Alegre
%S Proceedings
%K tomography, superpixel, rock, pores.
%X 4D micro and nanotomography allows the study of time-resolved phenomena, such as understanding how water, oil, and gas interact within rock pore space to improve oil engineering. This category of experiment has only become possible due to the invention of powerful devices such as the MOGNO micro and nano-tomography beamline of Sirius, the new state-of-the-art Brazilian synchrotron light source. Nowadays, the biggest bottleneck for data analysis in this type of experiment is the image segmentation task, given that MOGNO may generate one 3.6 gigavoxels 3D image in 1-5s. To achieve near real-time image segmentation in the future and reduce manual processing, we propose to convert the image segmentation task into superpixel classification. We have evaluated different combinations of superpixel estimation algorithms, feature extraction filters, and pattern classifiers aiming to automatically segment the pore space in 3D micro-CT rock grain images.
%@language en
%3 SIBGRAPI_WIP_2018.pdf