Abstract: The Yellow River Basin is an important distribution area of mineral resources in China.For a long time,high-intensity mining in the mining area has caused a series of problems such as soil erosion and vegetation degradation.As the energy power of the mining ecosystem,vegetation has great carbon fixation rate and potentiality.Therefore,accurate acquisition and monitoring the growth state of vegetation in the mining area is of great significance to the ecological protection and high-quality development of the Yellow River Basin.Proposes an automatic extraction method of vegetation in mining area based on UAV-based dense matching point cloud.The sequential images of mining area are obtained through the digital camera carried by UAV,then the three-dimensional point cloud in mining area is reconstructed through feature extraction,spatial three measurement,and multi view image dense matching.Using the rich spectral characteristics of dense matching point cloud,the differential vegetation index DEVI of point cloud is constructed,and the global threshold is automatically obtained through Otsu Method to realize the automatic extraction of vegetation point cloud.A mining area of the Yellow River Basin in Henan Province is selected for experiments.The results show that the overall accuracy of vegetation extraction is 96.40%,and the kappa coefficient is 0.927 1,which can effectively extract the three-dimensional information of vegetation from the UAV-based dense matching point cloud,and provide a feasible method for the study of three-dimensional monitoring of mining area vegetation based on low-cost UAV photogrammetry.