KLASIFIKASI DATA MINI LIDAR DENGAN WAHANA UAV PADA KAMPUS ITB JATINANGOR

• Main Author: Irham, Muhammad
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/51171
• Institution: Institut Teknologi Bandung

3D map is a map displaying data related to the place of an object horizontally and vertically. The need for 3D maps increases along with the development of science and technology in making these maps. LiDAR (Light Detection and Ranging) is one of the best technologies that can be used for 3D mapping. This technology uses light in the form of a laser to determine the distance of an object on the earth's surface. The LiDAR data acquisition process can use various ways and methods, one of which is by using an unmanned aircraft (UAV). The resulting data is a collection of point clouds which is coordinated to form objects and in a certain coordinate system. LiDAR data should be presenting clearer information from the differentiated objects, therefore a classification process is required. The purpose of this study is to determine the classification parameter, produce classified data by appropriate methods, and assess the accuracy and precision of the LiDAR data classification results. This research’s methodology comprises literature studies, data acquisition to data processing. The data processing consists of automatic classification using algorithms of the software, and manual classification for accuracy and precision testing purposes. The data used in this research is the point cloud of Kampus ITB Jatinangor. The object classes defined in this study are the classes of land, vegetations, buildings, and unclassified clouds. In the soil classification process, the Maximum Local Slope and Morphological Filter methods are used with the value of parameters determined by trial and error. The vegetation classification process using the Morphological Filter method with parameter values obtained from the LiDAR 360 manual book. In the building classification process, the Surface Growing method is used with parameters determined by trial and error. Based on the parameter values determined in this study, the grade of accuracy is 96.4% and the grade of precision is 0.964.