Flood Damage Assessment Geospatial Application Using Geoinformatics and Deep Learning Classification

Flood Damage Assessment Geospatial Application Using Geoinformatics and Deep Learning Classification

Flood damage assessment

Geoinformatics

Deep Learning Classification

The data of impacts and damage caused by floods is necessary for manipulation to assist and relieve those impacts in each area. The main issue for data acquisition was acquisition methods that affect the durations, accuracy, and completeness of data obtained. Most data are currently obtained by field survey for data on impacts in each area. However, this method contains limitations, i.e., taking a long time, high cost, and no real-time data visualization. Thus, this research presented the study to develop an application for inspecting areas under impact and damage caused by floods using deep learning classification for flood classification and land use type classification in the affected areas using digital images, remote sensing data, and crowdsource data notified by users through the accuracy assessment application of classification. It was found that deep learning classification for flood classification had 97.50% accuracy, with Kappa = 0.95. Land use type classification had 93.71% accuracy, with Kappa = 0.91. Flood damage assessment process in this research was different from other previous research that used geospatial data for flood damage inspection, e.g., satellite images. In contrast, this research brought damage data notified by users for processing with flood data in each area by satellite image processing and land use types of classification. The proposed application can calculate damage in each area and visualize real-time results in maps and graphs on the dashboard via the application. Besides, the presented method can be used to verify and visualize data of areas under impact and damage caused by floods in different areas.

Puttinaovarat, Supattra

Saeliw, Aekarat

Pruitikanee, Siwipa

Kongcharoen, Jinda

Chai-Arayalert, Supaporn

Khaimook, Kanit

International Journal of Interactive Mobile Technologies (iJIM); Vol. 16 No. 21 (2022); pp. 71-97

1865-7923

International Association of Online Engineering (IAOE), Vienna, Austria

2022-11-15

Copyright (c) 2022 Assoc.Prof.Dr.Supattra Puttinaovarat, Aekarat Saeliw, Siwipa Pruitikanee, Asst.Prof.Dr.Jinda Kongcharoen, Asst.Prof.Dr.Supaporn Chai-Arayalert, Assoc.Prof.Dr.Kanit Khaimook

https://creativecommons.org/licenses/by/4.0

application/pdf

eng

info:eu-repo/semantics/article

info:eu-repo/semantics/publishedVersion

Peer-reviewed Article