Automatic Detection of Mining Subsidence using InSAR and YOLOv11 Model

Abstract. Underground coal mining will cause serious ground subsidence and affect the life and property safety of surrounding residents. Deep learning provides the possibility to process a large amount of data information, and can provide an effective model based on a large number of training data to realize the automatic detection and recognition of a large number of data. In this paper, the YOLOV11 model is applied and a mining area in Shanxi Province was used for experimental research. The experiments show that an AI model suitable for automatic identification of subsidence area in Shanxi mining area is obtained, by training YOlOv11 model with limited SAR interferogram. The YOLOv11 model will also be applied to surface deformation caused by landslides and earthquakes in the future, so as to improve the identification efficiency of geological disasters and reduce the possible human and property losses caused by ground disasters.