Abstract: The Small Baseline Subset InSAR (SBAS-InSAR) and unmanned aerial vehicles (UAVs) as common ocean-land technologies, have been extensively applied in subsidence, glacial movement, surface deformation, and maritime positioning and navigation. A novel method integrating SBAS-InSAR and UAV photogrammetry is used to analyze ground subsidence deformation in the Gesar gold mine located in Maqu, Northwest China. This approach uses SBAS-InSAR to calculate two-dimensional deformation data for capturing ascending and descending measurements. This method can provide precise information on small-sized deformations within mining regions. The deformation data obtained from UAVs and the vertical deformation data derived from InSAR are integrated to generate comprehensive and accurate ground subsidence data from the mining district. Results demonstrate that using a combined InSAR (vertical) and UAV technique to analyze surface subsidence in mining districts resolves inconsistency between the line-of-sight and deformation orientations. Furthermore, the incoherence issue of InSAR in regions with large deformation gradients is addressed, while the inherent errors of UAV monitoring of mining surface subsidence are mitigated. The genetic algorithm (GA)-backpropagation (BP) neural network algorithm is combined with InSAR data to predict subsidence in collapsed areas. As observed, the GA-BP algorithm has the smallest residual under the same training samples. Therefore, the GA-BP neural network model can effectively predict surface subsidence in mining areas and can be used for subsequent subsidence prediction.