Abstract: To achieve high-quality panoramic visualization of deep wellbore interior morphology and overcome the limitations of traditional image stitching methods in terms of automation, stitching quality, and processing efficiency, an intelligent image stitching method integrating YOLOv11s and Structural Similarity Index Method is proposed. The research aims to address the challenges of high dependency on manual intervention, seam generation, uneven brightness, and visual artifacts inherent in existing technologies, thereby improving the accuracy, continuity, and overall efficiency of borehole image stitching. In terms of technical methodology, a lightweight YOLOv11s object detection network is introduced, making full use of its deep feature extraction capability and multi-scale detection advantages to accurately identify circular boundaries in borehole images. The center coordinates and radius parameters of the circles are automatically extracted, effectively overcoming positioning deviations caused by lens distortion, uneven lighting, or local occlusion. Subsequently, based on the precise geometric parameters, polar coordinate transformation is applied to unroll the circular inner wall area into rectangular images, preserving the original texture information while constructing a spatially ordered unfolding atlas. Building on this, the SSIM structural similarity measure is innovatively combined with a sliding window matching strategy. By systematically analyzing the brightness, contrast, and structural consistency of adjacent unfolded images in overlapping regions, the optimal registration position is adaptively searched, achieving efficient, seamless image stitching. This process ultimately generates a complete panoramic image of the borehole’s inner wall. Experimental results show that the proposed method takes only 34.98 seconds to stitch 240 images, significantly outperforming the traditional SIFT feature point matching method, which takes 271.35 seconds under the same experimental conditions. Additionally, the stitching results exhibit higher visual coherence and geometric fidelity, effectively suppressing common issues such as seam misalignment, brightness jumps, and texture repetition. The innovation lies in proposing an automated stitching workflow that combines YOLOv11s and SSIM algorithms to enhance stitching efficiency and visual quality.