Objectives Surface deformation monitoring is of great significance for deeply studying disaster formation mechanisms and evolution characteristics, as well as for establishing an integrated risk-based early warning system.

Methods To address the issues of low efficiency and limited accuracy in manually identifying ground control points in traditional photogrammetric methods, a high-precision method for extracting surface deformation using unmanned aerial vehicle (UAV) images was proposed. First, a precise automatic identification algorithm was developed to obtain the image coordinates of coded targets within UAV images based on the characteristics of coded targets. Then, the deformation results were generated quickly and accurately by using the image coordinates of coded targets in the UAV images captured before and after deformation and the object space coordinates of the control points. Finally, total station survey data were used as reference values to compare and verify the proposed method.

Results The experiment results of UAV images validated that the proposed method can achieve sub-centimeter accuracy. The precision of the proposed method is significantly higher than that of both the cloud-to-cloud comparison algorithm and the multi-scale model-to-model cloud comparison algorithm.

Conclusions The proposed method can accurately extract surface deformation data, indicating strong practicality and significant application potential.