Inventory of the quantitative characteristics of single oak trees with nonparametric methods of Support Vector Machines and Decision Tree on satellite images of WorldView-2 and UAV

In Iran s forests, forest statistics information is necessary for land management because less than 10% of Iran is formed by forests. So forest information is necessary for forest management, such as calculating the number, diameter at breast height (DBH), and volume. While traditional data is obtained using labor costs and time for terrestrial measurements, new technologies such as remote sensing provide us with up-todate data. Although many sensors extract the forest information for us, the satellite WV- 2 has very high resolution images. In the present study, we evaluated the estimation of forest parameters by focusing on single tree extraction by two methods of decision tree and Support Vector Machines classification with complex matrix evaluation and Area under operating characteristic curve (AUC) method with the help of UAV Phantom 4 Pro images in two distinct regions. The method of Support Vector Machines classification has the highest accuracy in estimating single tree parameters and then is decision tree method. This study confirms that using WV-2 data we can extract the parameters of single trees in the forest.