Single tree crown detection and segmentation method based on unmanned aerial vehicle image and U-Net

The invention provides an individual tree crown detection and segmentation method based on unmanned aerial vehicle images and U-Net, and relates to the technical field of individual tree segmentation. The method comprises the following steps: collecting unmanned aerial vehicle image data of multiple tree species in different scenes and making an orthoimage; the method comprises the following steps: intercepting orthoimage sub-images corresponding to different scenes, processing the orthoimage sub-images to obtain required sample data, and performing data enhancement on the sample data to obtain a sample data set; dividing the sample data set into a training set and a test set; obtaining a U-Net model, training and verifying the U-Net model by using the training set and the test set, supervising a prediction result of the U-Net model trained each time by using the mixed Loss function, stopping training of the U-Net model until the mixed Loss function is stable and does not descend any more, and obtaining a tra