Classification Mapping and Statistics Based on UAV Image Urban Buildings
Buildings are the main places of human life,and classification and statistics of them can provide the necessary basic data and information for urban planning,resource management,disaster risk assessment,environmental monitoring,and other important matter,thereby supporting the formulation of correct decisions and the sustainable development of cities.To solve the problem of small spacing between buildings in remote sensing images,a classification mapping and statistics of UAV image buildings in Anqing urban area are carried out based on the random forest algorithm in machine learning and multi-scale segmentation.The results show that the classification accuracy of this method is 0.873 4,and the Kappa coefficient is 0.762 7.Through comparative analysis using visual interpretation,it is found that this method is feasible for building classification.
国家科技期刊平台
NSTL
万方数据
