CT&F - Ciencia, Tecnología y Futuro
versão impressa ISSN 0122-5383
COBALEDA, G. E. A STUDY OF IMAGE AUTOCORRELATION AS A METHOD TO CHARACTERIZE POROSITY AND ESTIMATE PERMEABILITY. C.T.F Cienc. Tecnol. Futuro [online]. 2000, vol.2, n.1, pp. 69-81. ISSN 0122-5383.
Ioannidis, et al., published a statistical methodology for analyzing of porous microstructure to estimate permeability in the Journal of Petroleum Science & Engineering. The method is based on the calculation of the autocorrelation function from binary images of porous space. This paper presents a systematic application of the methodology applied to twenty-three sandstone samples from a Colombian oil bearing formation. Images of porosity were acquired under different detection conditions in order to carry out a comparative evaluation of the results at different resolutions. It was found that the adjustment parameters for the autocorrelation function, as well as its derivative and integral, dependent strongly upon the conditions of image detection. Under conditions of maximum resolution, permeability was estimated within the same order of magnitude of the value measured experimentally, with an accuracy that varies from 1.6% to 24%.
Palavras-chave : autocorrelation; length of correlation; permeability; pore geometry.