DocumentCode
2956573
Title
Fractal analysis of multiscale pore connectivity networks
Author
Lian, Teo Lay ; Sagar, B. S Daya
Author_Institution
Fac. of Eng. & Technol., Multimedia Univ., Melaka, Malaysia
Volume
2
fYear
2003
fDate
18-20 Sept. 2003
Firstpage
686
Abstract
Mathematical morphology method is applied to extract physically-realistic of pore connectivity network (PCNs) from a sandstone microphotograph. The fractal dimensions that have been estimated by box dimension method which have precisely followed the universal power-law relationships proposed elsewhere. These results are derived by considering structuring elements such as octagon, square and rhombus which have been used to extract pore connectivity network of pore-space of sandstone image. The structuring elements can be used to correlate pore scales phenomena and can also be incorporated into a pore space to verify the existing connectivity network in pore space. From the number-size relationship, the fractal dimensions of pore-space estimated, by considering these structuring elements, yield the values of 1.67, 1.68, and 1.69. These values were confirmed with the values arouse from our previous paper, estimation of morphological decomposition method.
Keywords
feature extraction; fractals; image processing; mathematical morphology; microphotography; porous materials; box dimension method; feature extraction; fractal dimension analysis; image processing; mathematical morphology; multiscale pore connectivity network extraction; pore scale phenomena; pore space estimation; porous materials; sandstone microphotograph; structure element; universal power-law relationship; Electronic mail; Extraterrestrial phenomena; Fractals; Geometry; Image analysis; Morphology; Personal communication networks; Shape; Solid modeling; Yield estimation;
fLanguage
English
Publisher
ieee
Conference_Titel
Image and Signal Processing and Analysis, 2003. ISPA 2003. Proceedings of the 3rd International Symposium on
Print_ISBN
953-184-061-X
Type
conf
DOI
10.1109/ISPA.2003.1296363
Filename
1296363
Link To Document