This paper addresses the limitations of traditional dam safety monitoring methods, which are characterized by low efficiency, high cost, and limited coverage. A novel dam inspection method based on close-range photogrammetry technology is proposed. By employing high-resolution cameras mounted on drones for close-range photogrammetry, combined with computer vision and deep learning algorithms, this method achieves high-precision detection and quantitative analysis of surface cracks, seepage, deformation, and other defects on dams. The study conducted experiments on three dams of different types, and the results demonstrated that the proposed method achieved a crack detection accuracy of ±0.1 mm and a deformation monitoring accuracy of ±1.2 mm. Compared with traditional methods, the efficiency was improved by 5 to 8 times, and the cost was reduced by over 60%. This research provides an efficient, precise, and cost-effective innovative solution for dam safety monitoring and holds significant importance for promoting the intelligent inspection of water conservancy projects.