Classification of UAV point clouds by random forest machine learning algorithm

Today, unmanned aerial vehicle (UAV)-based images have become an important data sources for researchers who deals with mapping from various disciplines on photogrammetry and remote sensing. Reconstruction of area three-dimensional (3D) point clouds UAV-based are essential process to be used traditional 2D cadastral maps or produce a topographic maps. Point should classified since they subjected analyses extraction further information direct cloud data. Due the high density clouds, processing gathering makes classification challenging task may take long time. Therefore, allows optimal solution acquire valuable information. In this study, random forest machine learning algorithm is applied radiometric features (Red band, Green band Blue band) geometric characteristics derived covariance feature (curvature, omnivariance, flatness, linearity, surface variance, anisotropy normalized terrain surface) points. addition, case study presented in order test applicability proposed methodology accuracy performance method UAV based cloud. After processing, class assigned each model was compared reference class. Lastly, overall achieved as 96% Kappa index reached 91% set.