An IoT Enabled Knowledge Graph Convolutional Radial Basis Deeplabv3+ Model for Flood Disaster Detection

Shashank Shashikant Tolye (), Vinayak Ashok Bharadi () and Kaushal Kamaleshwar Prasad ()
Additional contact information
Shashank Shashikant Tolye: Finolex Academy of Management & Technology, Department of Information Technology
Vinayak Ashok Bharadi: Finolex Academy of Management & Technology, Department of Information Technology
Kaushal Kamaleshwar Prasad: Finolex Academy of Management & Technology, Department of Mechanical Engineering

Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA), 2025, vol. 39, issue 14, No 9, 7571 pages

Abstract: Abstract Most prevalent and destructive natural disasters represent all weather-related incidents and impact more individuals, especially during floods. These regions assess the susceptibility to flooding, which is crucial for effective disaster risk mitigation. However, current methods for estimating flood damage, including depth-damage functions, frequently fall short regarding regional relevance and precision. To tackle these challenges, an innovative approach known as Knowledge Graph Convolutional Radial Basis DeepLab v3+ (KGCRDLv3) has been developed in conjunction with the Arctic Fox Optimization Algorithm (AFOA) to improve the precision and efficiency of flood zone identification. This method integrates the LIME is utilized in the KGCRDLv3-AFOA methodology, in order to afford interpretable and local explanations of flood zone predictions, which provide the most significant spectral and graphical features in satellite images and knowledge graphs to help make the model more transparent and trustworthy. The incorporation of knowledge graphs within KGCRDLv3 enhances both spatial and contextual comprehension. At the same time, the radial basis convolutional layers are proficient in recognizing complex patterns and addressing fine details in satellite imagery. Additionally, a Long-Range Wide Area Network (LoRaWAN) communication network interface enables the transfer of collected data to cloud servers through the internet of things-enabled devices like Raspberry or Arduino. This method ensures low power consumption and improves scalability in data communications. The proposed approach achieves an accuracy of 99.7%, precision of 98.8%, recall of 98.5%, specificity of 99.2%, detection rate of 98.3%, area under curve of 0.98 and prediction time of 10.52 s. Overall, the suggested approach enhances flood area detection by improving spatial-contextual understanding and pattern recognition.

Keywords: Natural Disasters; Flood Susceptibility; Disaster Risk Mitigation; Flood Damage Estimation; Low Power Consumption (search for similar items in EconPapers)
Date: 2025
References: Add references at CitEc
Citations:

Downloads: (external link)
http://link.springer.com/10.1007/s11269-025-04307-2 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:waterr:v:39:y:2025:i:14:d:10.1007_s11269-025-04307-2

Ordering information: This journal article can be ordered from
http://www.springer.com/economics/journal/11269

DOI: 10.1007/s11269-025-04307-2

Access Statistics for this article

Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA) is currently edited by G. Tsakiris

More articles in Water Resources Management: An International Journal, Published for the European Water Resources Association (EWRA) from Springer, European Water Resources Association (EWRA)
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().