Abstract: Quantitative characterization of rockmass discontinuities has significant impact on the evaluation of slope stability. Considering the situation of the complex terrain and the difficulty of manual measurement at high and steep rock slope,unsatisfactory slope model construction affects the accuracy of slope stability analysis. This paper presents a field application on Yinshan open-pit copper mine in Dexing,Jiangxi Province. 3D point cloud model is first generated based on unmanned aerial vehicle(UAV)photogrammetry integrated with a structure from-motion(SfM)technique. An automatic procedure for discontinuity detection,digital trace mapping and discontinuity parameters is established. The parameters include the orientation,number of sets,trace length,and set spacing. The extracted values are compared with planned manual measurements through scanline method. The results show that the slope shape and the rockmass model obtained by the UAV photogrammetry are more in line with the actual rock slope. The discrete fracture network(DFN) is established based on the statistic model of rockmass. Then a 3D numerical model is quickly built using non-uniform rational B-spline(NURBS)surface reconstruction and a combined modeling method of "Rhino-Griddle-3DEC". And better numerical simulation results are obtained to meet the demands of open pit mine slope stability analysis.