ANALISIS KERUSAKAN BANGUNAN WILAYAH PESISIR PALU AKIBAT TSUNAMI MENGGUNAKAN DEEP LEARNING

• Main Author: Ikbal Rahadian, Achmad
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/48333
• Institution: Institut Teknologi Bandung

Indonesia has a quite high tsunami risk because it traversed by four tectonic plates, which are Philippine, Eurasian, Pacific and Indo-Australian plates. Considering the fact that Indonesia is an archipelagic country with two-thirds of its territory is water and has the second longest coastline in the world, the potential for tsunami damage is enormous in Indonesian territory. Nonetheless infrastructure damage, especially building due to the tsunami hazard, is rarely assessed quantitatively, even though the assessment is the first step in the rehabilitation and reconstruction of a tsunamiaffected area, and is further used as a basis for policy making in the disaster mitigation process. The assessment can be done quickly without having to go directly to the location where the disaster occurred by conducting analysis using satellite imagery. In this study, an analysis of building damage due to the tsunami was carried out with the study location of the Palu coastal area which experienced a tsunami on Friday, September 28, 2018 at 18:02 WITA due to an earthquake with magnitude of 7.5. In the analysis of building damage due to the tsunami, the location of the distribution and the number of buildings were generated through object detection using deep learning from high resolution satellite imagery data. Object detection was carried out using pretrained YOLOv3 models that are trained using 315 satellite images as data sets and produce a model with a loss value of 33.15. Object detection was carried out on satellite imagery before and after the tsunami and produced building distribution data with an accuracy of 76.78% and 74.20%, respectively. Comparison of building data detected from the two satellite images then analyzed using a tsunami height zone map to see the correlation between building damage and tsunami height. Research result show there is 1,547 damaged buildings were detected and shows a positive linear correlation between the tsunami height variable and the building damage variable. Using the t-student test it was concluded that there is a significant relationship between both variables.