The measurement of rock mass structural plane orientation is a critical step in field geological investigations, as the morphology, properties, and orientation of structural planes are key factors influencing rock mass characteristics. This study employed widely used UAV photogrammetry to obtain a 3D digital model of slope rock masses. Key structural planes were manually identified, and the coordinates of three or more points on each plane were selected to fit a plane equation using the least squares method. The normal vector of the plane was then derived to calculate the orientation of the structural plane. For structural planes only exposed as traces on the surface, multiple points along the polyline of the trace were extracted to fit the plane equation, and the orientation was determined using the same method. To verify the reliability of the approach, the accuracy of the orientation calculation program was first tested on a sloping roof (dip direction error: 0.47°) and a flat roof (dip angle error: 0.31°), both with errors of less than 1°. On well-exposed structural planes at the Zhengjiatai and Hongniangkou slopes, comparisons between calculated and field-measured results show absolute dip direction errors ≤ 3.34°, absolute dip angle errors ≤ 1.67°, and relative errors all below 2.5%, indicating good agreement. For structural planes only exhibiting traces on the Xitaiping Road and Xishimen slopes, the fitted orientations based on trace point coordinates were compared with measured values (dip direction error ≤ 3.07°, dip angle error ≤ 0.77°). Additionally, comparisons between trace points and surface points on damaged concrete pavement (dip direction difference: -0.94°, dip angle difference: -0.05°) further validate the method’s effectiveness, and address the challenge of determining orientations for trace-only structural planes. The process of fitting a plane using multi-point spatial coordinates, deriving the normal vector, and calculating structural plane orientation was implemented in a Python program, which has achieved rapid, accurate, and batch acquisition of orientations. This approach resolves technical challenges such as cumbersome orientation calculation procedures, difficulties in obtaining continuous orientations, and challenges in measuring orientations on high and steep slopes, indicative of its strong practical applicability.