Flood Disaster Identification and Decision Support System using Crowdsource Data Based on Convolutional Neural Network and 3S Technology

Supattra Puttinaovarat, Teerawad Sriklin, Sompop Dangtia, Kanit Khaimook


Every flood causes damages to many lives and properties. Moreover, it affects the economy and lifestyle of people in the society in a short period and a long period. In consequence, this research would demonstrate techniques and flood detection analyzed through digital images and web application development for receiving reports and inspection of flood situation in every area. The process requires crowdsource data and uses 3S technology so it will receive the accurate and real-time data for making a decision in the flood management and aids to the people in the disastrous area. In this research, Convolutional neural network was applied for flood detection and classification using digital images and data from people or the victims. According to the study, it was found that convolutional neural network for flood classification has accuracy of data at the high level or 95.50%, 93.00%, 97.89%, and 0.91 which are the results of accuracy, producer accuracy, user accuracy, and kappa statistics, respectively. Besides, the use of this technique saves cost, time, and labors. Furthermore, the method could be applied to other disasters such as landslide, earthquake, and fire. It is able to monitor the incident in each type of disasters and also examines the damaged site after the incident.


Flood Identification; CNN; Crowdsource Data; GIS

Full Text:


International Journal of Interactive Mobile Technologies (iJIM) – eISSN: 1865-7923
Creative Commons License
Scopus logo IET Inspec logo DBLP logo EBSCO logo Ulrich's logo MAS logo