Detection of rob flood affected areas in Belawan with medium resolution imagery using deep learning.

Based on Badan Nasional Penanggulangan Bencana (BNPB), December 28, 2021, throughout 2021 from January 1 to December 28, 2021, natural disasters that occurred in Indonesia reached 3,058 events. Flood disasters dominate natural disasters that hit almost all parts of Indonesia, reaching 1,288 events or 42.1% [6]. this number is not a small number but quite large, of course this has a very bad impact on human survival in the future. Flooding is a condition in which an area is inundated by large amounts of water. This disaster is usually caused by an increase in the volume of water in a water reservoir, such as rivers, lakes, seas, etc., resulting in the overflow of water dams from their natural limits that submerge the land. Rob flood is an event of rising sea water caused by tidal activity of sea water that inundates the coastal land or a place lower than the highwater level. Tidal flooding is a threat to people in coastal areas because it can cause a lot of harm to the social and economic life of the local community. The puddle of water caused by the tidal flood occurred due to land use and topography that were no longer able to absorb water. The distribution of puddles due to the Rob flood that is not immediately resolved can cause losses to the affected area, so to prevent this from happening it is necessary to conduct an in-depth study to detect areas affected by the Rob flood by using Deep Learning. Deep learning is one of the implementations of Machine Learning which aims to imitate the workings of the human brain using an Artificial Neural Network. Deep Learning with a number of algorithms as "neurons" will work together in determining and digesting certain characteristics in a data set. Programs in Deep Learning usually use more complex capabilities in studying, digesting, and also classifying data. In this study, researchers used the KTrain and ResNet-50 method. The expected output of this research is the establishment of a reliable system to detect the Rob area as a form of handling the Rob flood disaster. [ABSTRACT FROM AUTHOR]