To tackle the optimization challenge of image segmentation in three-dimensional digital core construction via computed tomography (CT) scanning, systematic comparisons of threshold-based and morphological image segmentation features were conducted using the Avizo software platform. Their applicable contexts were clarified accordingly.Microscopic pore analysis of sandstone samples from the Putaohua oil layer (Yaoyi Member, Daqing Oilfield) demonstrates that micropores with an equivalent diameter <30 μm constitute 90% of the total pore frequency but contribute less than 20% to pore volume. In contrast, pores with an equivalent diameter>30 μm, though representing merely 10% of the frequency, account for over 75% of the volume—confirming them as the predominant reservoir space.Seepage modeling based on the lattice Boltzmann method indicates that the primary controlling factors of permeability are tortuosity, coordination number, and throat radius in descending order. The fractal dimension only quantifies pore structure complexity and bears no notable correlation with permeability.A nonlinear fitting curve between absolute horizontal permeability and tortuosity was developed, with validation carried out using 25 mm core plugs. These findings offer a theoretical foundation for the selection of digital core image processing technologies and the prediction of reservoir seepage behaviors.