Objective Gaps between the electrodes of formation microresistivity imaging (FMI) imagers lead to blank strips in the resistivity images of borehole walls, significantly influencing the parameter assessment for fractures near borehole walls. The absence of full-borehole images renders it challenging to assess the blank strip filling quality for fractures in FMI images. Using a dataset constructed utilizing both simulated and actual data, this study proposed a generative adversarial network (GAN)-based method for filling the blank strips of fractures in FMI images.

Methods First, the resistivity logging responses of a fractured formation were simulated using the 3D finite element method. Actual fractures were simplified using the transition boundary condition, improving the computational efficiency by avoiding multi-scale models and meeting the requirements of deep learning for substantial samples. Second, full-borehole images were attained by integrating the simulated fractures with actual FMI images, and masks generated from FMI images were used as blank strips. To achieve the optimal filling effects, hyperparameters were optimized by assessing the capacity of blank strip filling for fractures using image filling indicators. Finally, the blank strips of fractures were filled using masks with varying proportions, followed by a filling quality assessment.

Results and Conclusions  The results indicate that the proposed method allows for the filling of blank strips with varying proportions. Notably, it yields encouraging filling effects and smooth morphology for coarse fractures and can restore their margins and details more accurately. Its applicability was further verified using actual log data, revealing that it produced natural-looking filled images that can effectively restore fracture features. The proposed method facilitates the extraction and quantitative analysis of fractures, laying a foundation for the precise assessment of fractured reservoirs and supporting the accurate prediction of hydrocarbon productivity.