#TITLE_ALTERNATIVE#

• Main Author: ROFIATUL AINIYAH (NIM: 15109090); pembimbing: Agus Suparman, Dr., Ir., M.Sc ; Budhy Soeksmantono, S
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/19362
• Institution: Institut Teknologi Bandung

Light Detection And Ranging (LiDAR) is a remote sensing technology using aircraft equipped with sensors to emit laser pulses, the distance between the sensor and the object on the surface of the earth is calculated by recording the time it takes for a laser pulse is emitted and returned to the sensor. Point clouds are LiDAR data which can be a collection point of information containing x, y coordinates, and height (z). On the application of LiDAR point clouds are not all used, for example in the manufacture of Digital Terrain Model (DTM), the point clouds are used are ground points, while in the 3D modeling of an area of the point clouds are used is non-ground points (building points). For that needs to be done to the clouds points classification process required classes. The purpose of this study is to look at the appropriate classification method used. Classification process that is used to divide the points clouds into ground points and non-ground points, in this research is the automatic classification performed by using the algorithm in software and manual classification made by researcher if needed. The automated classification is done by performing classification points directly to the ground, and also automatic classification is done to some classes to get class ground points, the method is also carried out low-class classification to the class points to the points which are considered as noise . Manual classifications performed to classify the points that have not been properly classifiable. To see the results of the classification accuracy of the test by doing overlay ground points by classification points to the orthophoto. <br /> <br /> <br /> <br /> <br /> <br /> <br /> Based on the classification results of the two methods of automatic classification is performed, the classification points directly to the ground still contains the points that a noise while the automatic classification into some classes have eliminated the noise on grounds points. In the overlay of the test results still missing ground points are located on the roof of the building. The results of overlay ground points by classification with orthophoto showed that in areas where there is a regular building ground points at the top of the building is less than the area of the irregular <br /> <br /> <br /> <br /> <br /> <br /> <br /> buildings. As well as the trial results using the parameter values for the different classification algorithms, classification with small terrain angle is better than bigger terrain angle..