砂岩顆粒間的孔隙可作為能源封存或儲藏空間，其孔隙率決定儲藏油氣等資源的潛能，孔隙幾何形狀與滲透率則影響能源開採難易度。近年X光電腦斷層掃描(XCT)提供了有效的方式，做為非破壞性的砂岩內部結構分析。在電腦斷層影像進行影像分割時，影像灰階直方圖之門檻值是判讀影像固體及孔隙間極重要之參數。調整門檻值標準，將影響孔隙幾何形狀、分佈及孔隙率之判讀。本研究運用不同影像處理方法進行巨量X光電腦斷層切片處理，後續分析澳洲砂岩及Darley-Dale砂岩孔隙率。實驗樣品首進行X光電腦斷層掃描，並完成二維平面切片影像重建。將獲得灰階影像，以監督式學習（類神經網路演算法）及非監督式學習（區域成長法及門檻值分割法），進行孔隙與固體相分割，並比較三種影像處理方法優劣勢。其中類神經網路演算法輸出與目標均方誤差值設定為10-5，進行倒傳遞網路疊代計算。經計算Darley-Dale砂岩（解析度9μm）孔隙率為13.0％；另澳洲砂岩孔隙率為13.1％（解析度9μm）及16.1％(解析度0.6μm)，此非破壞性之電腦斷層檢取代傳統汞滲法，有助於砂岩內部結構探討，並維持樣品完整性。

Porosity is an important property of sandstone owing to the underground resource storage available in this intergranular space. Recently, X-ray computed tomography (XCT) has been widely used to reconstruct the inner structure of rock samples in a non-destructive way. In this study, we employed XCT to investigate the porosity of Darley Dale and Australian sandstone. The large quantity of sliced images on CT was processed using various algorithms in this study and then the porosity for Darley-Dale sandstone and Australian sandstone was discussed after the image processing. 2D (dimensional) image post-processing is a necessary procedure before the reconstruction of a 3D model. Meanwhile, three image processing methodologies, threshold value segmentation, the region growth method and a neural network, were used to distinguish between the voids and solid part of the 2D images. The results demonstrated that the porosity ratio of Darley Dale sandstone is 13.0% using different computing algorithms, which is well-matched to pervious studies. Furthermore, the porosities of Australian sandstone (using the images with the resolution of 9μm and 0.6μm) were found to be 13.1% and 16.1%, respectively, under various image resolutions. Combining the porosity results and based on the above referenced measurements, we conclude that Australian sandstone demonstrated more porous behavior as compared to Darley Dale sandstone.

1.Appoloni C.R., Fernan C.P., Rodrigues C.R.O., 2007, X-ray micro tomography study of a sandstone reservoir rock. Nuclear Instruments and Methods in Physics Research A, V. 580 , p629–632 .
2.Cortina-Januchs M.G., J., Vega-Corona A., Tarquis A.M., Andina D., 2011, Detection of pore space in CT soil images using artificia neural networks, Biogeosciences, v.8, p.279–288.
3.Dickinson, W. R., 1983, Interpreting provenance relations from detrital modes of sandstones.
4.Dupuy P.M., Austin P., Delaney G.W., Schwarz M.P., 2011, Pore scale definition and computation from tomography data., v.182, p2249–2258.
5.Gonzalez R. C. and Woods R.E., 2001, Digital Image Processing, Prentice Hall, p.612-615.
6.Geet M.V., Swennen R., Wevers M., 200, Towards 3-D petrography: application of microfocus computer tomographyin geological science, Computers & Geosciences V.27 p.1091–1099.
7.Heap M.J., Baud P., Meredith P.G., Bell A. F.,Main I.G., 2009, Time-dependent brittle creep in Darley Dale sandstone. J. GEOPHYSICAL RESEARCH, V.114, B07203, P.1-22.
8.Jasti J. K., Jesion G., Feldkamp L.,1993, Microscopic Imaging of Porous Media With X-Ray Computer Tomography. September.p189-193.
9.Jiang H., 2009, computed tomography principles, design, and recent advances, John Wiley & Sons, Inc, p.37-40.
10.Ketcham R.A., Carlson W.D., 2001, Acquisition, optimization and interpretation of X-ray computed tomographic imagery: applications to the geosciences, Computers & Geoscience, vol 27, pp 381-400.
11.Kung J., Chien Y.H., Wu W.J., Dong J.J., Chang Y.T., 2012 ,Comparative sound velocity measurements between porous rock and fully-dense material under crustal condition: The cases of Darley Dale sandstone and copper block, American Geophysical Union Fall Meeting 2012, T23D-2700.
12.Nauck D., Klawonn F., Kruse R., 1997, Foundation of Neuro-Fuzzy systems, John Wiley & Sons, Inc, p.10-15.
13.Promentilla M. A. B. and Sugiyama T., 2010, Application of Microfocus X-ray CT to Investigate the Frost-induced Damage Process in Cement-based Materials, Advances in Computed Tomography for Geomaterials: GeoX 20 10, pp.125.
14.Vergés E., Tost D., Ayala D.,. Ramos E, Gr S., 2011, 3D pore analysis of sedimentary rocks., Sedimentary Geology , v.234 ,p.109–115.
15.YANG B.H., WU A.X., MIAO X. X., LIU J.Z., 2014, 3D characterization and analysis of pore structure of packed ore particle beds based on computed tomography image. Trans. Nonferrous Met. Soc. China, V.24, P.833-838.
16.簡佑祥，2012，砂岩彈性行為與其組織結構之研究 碩士:國立成功大學 p60-62
17.繆紹綱，2010，數位影像處理活用matlab，p5-15。