High-precision in-situ acquisition of joint morphology data and geometric heterogeneity study of roughness

The joint roughness coefficient (JRC) is a key parameter for evaluating the shear strength of rock masses and the stability of rock slopes. However, obtaining high-precision in situ joint surface morphology data on steep rock slopes remains challenging. This study proposes a UAV-based multi-angle nap-of-the-object photogrammetric method, which enables vertical imaging of joint surfaces by flying at close range and adjusting the shooting angle, allowing accurate acquisition of 3D joint morphology in the field. The method was applied to a high-steep slope on the left bank of the Sequ River in Tibet, where a 3D point cloud model with a resolution of 7 mm was constructed. Forty-nine joint samples larger than 2 m² were extracted and expanded to 1176 analysis samples through scale magnification and shear direction variation. Roughness analysis based on the θ*_max/(C + 1)^3D parameter shows that joint roughness approximately follows a log-normal distribution at small scales but gradually deviates as scale increases; moreover, roughness decreases exponentially with increasing point interval. Anisotropy analysis reveals that directional variation in roughness diminishes with growing scale, and the anisotropy ratio approximately follows a normal distribution. The results demonstrate that this multi-angle photogrammetric technique effectively overcomes technical constraints in complex terrain, providing a reliable data foundation and methodological support for the quantitative estimation of JRC and slope stability evaluation in high-steep rock slopes.