Landscape-level vegetation classification and fractional woody and herbaceous vegetation cover estimation over the dryland ecosystems by unmanned aerial vehicle platform

•The new method to map different vegetations and estimate their fraction vegetation cover on drylands by UAV was developed.•The new method is more efficient than three popular machine learning algorithms (SVM, RF, GTB).•The new method was developed into an interactive web-based software UAV-HiRAP.•A new model was developed to optimize the workload of vegetation investigation.