The morphology and connectivity of subsurface fracture networks are critical factors controlling wellbore stability and hydraulic fracturing efficiency. Accurate characterization of the three-dimensional complexity of fractures holds significant importance for engineering safety and performance enhancement. A novel image segmentation model is established in this study. It enhances the iterative threshold method by incorporating simple linear iterative clustering superpixels, ResNet50, and a Gaussian mixture model. The model first divides complex computed tomography images into numerous superpixel images using simple linear iterative clustering superpixel segmentation. Subsequently, ResNet50 is employed to classify these superpixel images. Based on the classification results, the iterative threshold segmentation method is applied to segment different categories of superpixel images accordingly. Following preliminary image segmentation, Gaussian mixture module is used for denoising the segmented fracture images, resulting in high-precision segmented images. The two-dimensional segmented images are then reconstructed in three-dimensional space, and the three-dimensional distribution characteristics of fractures are analyzed. This study concludes that the new fracture segmentation method enables high-precision extraction of fracture regions. Compared with threshold segmentation, the morphological analysis noise value in the two-dimensional images segmented by the method proposed in this study was reduced from 0.21% to 0.08%. Fracture distribution in three-dimensional space is complex, and areas with larger fracture networks exhibit greater complexity in their three-dimensional distribution.

Shale hydration is a critical factor influencing wellbore stability. However, accurately quantifying hydration strain and tracking the dynamic migration of fluid imbibition fronts in tight matrices remains a significant challenge. To address this issue, we propose a high-spatial-resolution monitoring scheme based on distributed fiber optic sensing (DFOS). By applying a hydrophobic coating to alter the imbibition path, we established a one-dimensional (1D) unidirectional imbibition boundary condition and compared it with semi-immersed free imbibition. Representative shale specimens with varying geological characteristics were tested to capture the continuous spatiotemporal evolution of hydration strain under both conditions. Our results show that under the restricted 1D path, the onset of strain exhibits a regular sequential delay along the fluid invasion direction, successfully mapping the progressive advancement of the imbibition front. The comparison between specimens with and without hydrophobic coating confirms that the coating effectively alters the imbibition path, leading to pronounced spatial lag and amplitude attenuation due to gravity and viscous resistance. Furthermore, we identify a dual characteristic of hydration strain. On one hand, governed by intrinsic properties such as clay content and permeability, different specimens exhibit distinctly different strain morphologies. On the other hand, within the same specimen, strain profiles at different measurement locations display high temporal self-similarity and spatial similarity, reflecting the structural control of the rock matrix. These findings suggest that shale hydration is a highly heterogeneous process, and the morphological differences can, to some extent, reflect variations in rock structure. Overall, this study provides an important methodological reference for improving wellbore stability models and optimizing drilling fluid design in deep shale engineering.