Automatic identification of concealed dangerous rock blocks on high-steep slopes considering finite-sized discontinuity intersections

The identification and characterization of concealed in-situ rock blocks on high-steep slope exposures are critical in rock engineering, but remain challenging. This study employs advanced UAV-based photogrammetry to capture high-resolution discontinuity data and develops a fully automated rock block extraction method consisting of three steps: (1) determination of free face and non-free fracture intersections, (2) surface search for rock blocks on free face. and (3) extraction and analysis of rock blocks. This approach simplifies the determination of discontinuity intersections while maintaining high accuracy. By incorporating all types of discontinuities contributing to rock block formation, the method enables precise in-situ rock block identification and extraction. Application to a rock slope in China, produced results consistent with the rock blocks observed in the 3D model, highlighting its accuracy and practical value.