Vegetation functional diversity is an important indicator reflecting the carbon sequestration capacity of ecosystems.Estimating functional diversity is of great significance for evaluating the effectiveness of ecological restoration and guiding ecological restoration projects.However, it is difficult to estimate vegetation functional diversity with medium and low-resolution remote sensing data.To this end, this paper takes a coal mine dump in a semi-arid area as the test area, uses UAV remote sensing to obtain LiDAR and hyperspectral image data, inverts and reconstructs the functional diversity of vegetation communities, and reveals the differences of functional diversity under various restoration modes.The results show that the UAV can quickly obtain high-precision vegetation hyperspectral and laser point cloud data, and effectively estimate the morphological and physiological properties of vegetation communities.Based on morphological and physiological traits, the functional diversity of plant communities can be effectively and quantitatively described.The selection and configuration of vegetation in the initial stage of ecological restoration will have a significant impact on the functional diversity of the vegetation community after the succession.UAV remote sensing can quickly obtain fine community structure and function data, and provide a reliable basis for ecological restoration effectiveness evaluation and vegetation optimization.