| dc.description.abstract | Remote sensing is an alternative that offers several advantages due to the limitations in data collection. Direct identification of tree species requires a lot of time and effort. With drone image analysis, in addition to saving time, energy and costs, the data generated will be more detailed and reach the crown area. This study aims to map the University of North Sumatra arboretum area using UAV (Unmanned Aerial Vehicle) and identify tree species in the University of North Sumatra Arboretum using UAV (Unmanned Aerial Vehicle). The method used is OBIA (Object-Based Image Analysis) method based on the division of multiresolution areas into homogeneous areas with scale, shape, and compactness, using field verification points as many as 98 points which are tested using confusion matrix.The results of the study obtained 462 aerial photographs at an altitude of 100 m which produced an orthomosaic with a GSD (Ground Sampling Distance) of 2.23 cm/pix. The results of tree species identification obtained 12 types of Anthocephalus cadamba, Eucalyptus alba, Garcinia atroviridis, Spondias pinnata, Pometia pinnata, Samanea saman, Alstonia scholaris, Gmelina arborea, Terminalia catappa L, Paraserianthes falcataria, Gnetum gnemon, Nephelium lappaceum. The results of the accuracy test with the confusion matrix method using a scale of 100 showed an Overall accuracy value of 78.57% and with a scale of 150 showed an Overall accuracy value of 77.75%. The accuracy test results obtained show that the classification results are reliable. | en_US |
