Speaker
Description
Accurate image-based characterization of pore structure and permeability is often limited by the trade-off between field of view and resolution. To quantify how image resolution systematically biases pore metrics in dense sandstone, we construct a true-physical multiscale dataset by repeatedly scanning the same fixed surface region with scanning electron microscopy (SEM) and spatial correspondence across scales. Structural images were acquired at three pixel sizes: S1 (0.1 μm/pixel), S4 (0.05 μm/pixel), and S16 (0.025 μm/pixel), where the higher-resolution images tile the same physical area covered at lower resolution. A unified image-processing pipeline (denoising, contrast enhancement, and Yen adaptive thresholding) was applied to extract porosity, pore size distribution (PSD), pore-boundary roughness, fractal dimension, and permeability. As resolution increases, fine pores and boundary details become progressively resolved, leading to a clear increase in the identified porosity from 4.6% (S1) to 5.55% (S4) and 6.31% (S16). The PSD shifts toward smaller pores and becomes narrower at higher resolution, consistent with the decomposition of “artificially merged” pores observed in coarse images into multiple micropores at finer pixel sizes. Roughness and fractal dimension increase with resolution, indicating enhanced sensitivity to pore-boundary complexity and local heterogeneity. Permeability was estimated from the image-derived PSD under a capillary-bundle assumption using the Hagen–Poiseuille relation with porosity-based tortuosity correction. The inferred permeability decreases from 2.34×10-17 m2 (S1) to 1.77×10-17 m2 (S4) and 1.72×10-17 m2 (S16), with the magnitude of decrease diminishing at finer resolution, suggesting that overall permeability becomes effectively captured beyond a resolution threshold around 0.05 μm/pixel for this sample. The high-resolution estimates are in good agreement with the measured gas permeability (1.85×10-17 m2), supporting the reliability of the workflow when an appropriate resolution is selected. Overall, this study provides a physically registered multiscale SEM framework to quantify resolution-induced bias in pore statistics and permeability estimation, offering practical guidance for resolution selection in digital, image-based seepage analyses of dense sandstones.
| Country | China |
|---|---|
| Acceptance of the Terms & Conditions | Click here to agree |
