Abstract: Flexure structures are highly hazardous geological structures in underground coal mines. Their accurate identification and characterization have long been a technical challenge in the geological guarantee of the mines. Conventional seismic exploration methods struggle to achieve fine-scale identification of flexure structures due to their limitations in resolution and signal-to-noise ratio. In contrast, fully digital high-density 3D seismic exploration technology, characterized by wide frequency bands, wide azimuths, and high coverage frequency, can provide high-quality data for fine-scale interpretation of flexure structures. This study investigated a block in the Yangquan mining area. Using the fully digital, high-density 3D seismic exploration data, this study proposed and systematically implemented an offset vector tile (OVT)-domain split-offset coherence cube fusion technique. Specifically, the seismic data were classified into far-, moderate-, and near-offset intervals in the OVT domain. The coherence cubes in the three intervals were computed using the third-generation coherence (C3) algorithm. Furthermore, the attribute fusion technique was employed to integrate the split-offset coherence cube slices into a comprehensive coherence cube slice, enabling fine-scale characterization of the main flexure zone and other concealed structures within the study area. The results demonstrate that the identified flexure structures based on the proposed technique align with the actual wellbore exposures. By leveraging the rich azimuth and offset information in wide-azimuth seismic data, the proposed technique can effectively overcome the limitations of individual coherence cubes in terms of the multiplicity of solutions and resolution. Furthermore, by complementing and integrating seismic response characteristics across different offsets, this technique can significantly enhance the capability for fine-scale identification of flexure structures. Therefore, this study provides an effective technical approach for achieving transparent detection of concealed geological hazards in coal mines.