Abstract: UAV photogrammetry technology has been widely used in the field of mining subsidence monitoring. However, the existing methods are still difficult to meet the high standard requirements in terms of subsidence monitoring accuracy. Aiming at the urgent need to improve the accuracy of surface subsidence monitoring in UAV mining area,a spatial domain filtering algorithm of UAV point cloud based on mining deformation adaptation is proposed. Through deep fusion of mining subsidence deformation mechanism and image processing technology,the algorithm effectively suppresses the random error of DEM difference method in UAV photogrammetry,and constructs a point cloud spatial domain filtering method that conforms to the mining deformation characteristics of subsidence area. The feasibility and accuracy improvement effect of the algorithm are verified by simulation test field test and actual mining area engineering application. The experimental results show that the mean square error of UAV subsidence monitoring is reduced by more than 50% compared with the traditional DEM difference method in the simulated test field environment. In practical engineering applications,the measurement accuracy is improved by more than 38% compared with the conventional method,and the error of subsidence monitoring in engineering cases is controlled within 12 mm. This method significantly improves the accuracy of UAV subsidence monitoring,and provides reliable technical support for the engineering application of UAV deformation monitoring in mining areas. 

Key words: UAV photogrammetry,subsidence monitoring,adaptation to mining deformation,spatial domain,precision improvement