This study focuses on using UAV photogrammetry and point cloud data processing to extract and calculate key safety parameters for the dry beach of tailings ponds. UAV technology, known for its low cost, operational simplicity, and comprehensive data coverage, is particularly useful for surveying large and complex terrains. The research was conducted at the Baizhangzi Gold Mine in Liaoning, China, where UAVs were employed to capture high-resolution point cloud data of the tailings pond.Key techniques used include slope threshold filtering, Delaunay triangulation, and the alpha-shape method to extract features such as the dry beach boundary and crest points. These features are crucial for safety monitoring as they allow the calculation of parameters like slope gradient, beach length, and dam freeboard. The study emphasizes the precision of 3D point cloud data compared to 2D images, offering more accurate and reliable results for monitoring tailings dam safety. The results showed that the calculated parameters closely matched real-world measurements, with errors below 1%. This demonstrates the effectiveness of the proposed methods in ensuring the safety and stability of tailings ponds through regular monitoring. The research provides a scientific basis for future safety assessments and risk management in mining operations.