PEMANFAATAN DRONE UNTUK KLASIFIKASI TUTUPAN LAHAN KAWASAN HUTAN LINDUNG DI DESA LAMPO KECAMATAN BANAWA TENGAH KABUPATEN DONGGALA

Abstract

Land cover is the physical appearance of the earth's surface. Land cover can describe the relationship between natural processes and social processes. Land cover can provide very important information for modeling purposes and for understanding natural phenomena that occur on the earth's surface. The purpose of this study was to determine the effectiveness of the use of drones for land cover classification in the Protected Forest Area of ​​Lampo Village, Central Banawa District, Donggala Regency. This research was carried out for 3 months, from October to December 2021 by processing data acquisition using Drones, the image processing includes Geometric Correction, Manual Classification (on-screen digitization) and Accuracy Test (cappa accuracy and overall accuracy).The results showed that the RMSEr value from the Geometric Correction of the image was 0.583 m and the land cover consisted of Primary Forest with an area of ​​39.73 Ha, Secondary Forest 57.04 Ha, Shrub 1.27 Ha, Open Land 1.38 Ha, and Gardens. A mixture of 0.55 Ha with an accuracy rate of 96% (overall accuracy) and 95.05% (accuracy cappa).