Disaster Detection and Classification System using ML
Keywords:
Convolutional Neural Networks (CNN), Disaster prediction, Image analysis, Machine learning (ML), VGG16Abstract
Precise identification of disaster types and estimation of impact severity from imagery can significantly contribute to more effective disaster response policies. Machine learning (ML) algorithms have been proven to have excellent capability to read disaster imagery and draw relevant actionable conclusions in addressing the critical necessity for real-time intervention and mitigation. This research work advocates a system making use of ML methods for analysing visual information to identify different types of disasters namely earthquakes, floods, and wildfires and to measure the severity of damage. The methodology fuses cutting-edge techniques such as Convolutional Neural Networks (CNNs) for deriving important visual features, with regression procedures to analyse the intensity of impact. This automation enhances faster mobilisation of resources as well as fact-based prioritisation during rescue efforts. It also investigates the real-world deployment of the model, its limitations, and proposes future improvements to enhance performance in a wide range of disaster scenarios. Experimental results show the potential of image-based ML systems to revolutionise disaster assessment and inform more responsive emergency response structures.
References
CRED and UNDRR, “2020: The non-COVID year in disasters,” CRED, Brussels, 2021. Available: https://reliefweb.int/sites/reliefweb.int/files/resources/2020%20-%20the%20non-COVID%20Year%20in%20Disasters%20-%20Global%20Trends%20and%20Perspectives.pdf
N. Altay and W. G. Green III, “OR/MS research in disaster operations management,” European Journal of Operational Research, vol. 175, no. 1, pp. 475–493, Nov. 2006, doi: https://doi.org/10.1016/j.ejor.2005.05.016
H. S. Munawar, A. W. A. Hammad, and S. T. Waller, “A review on flood management technologies related to image processing and machine learning,” Automation in Construction, vol. 132, Dec. 2021, doi: https://doi.org/10.1016/j.autcon.2021.103916
D. D. Atsa’am, T. Gbaden, and R. Wario, “A machine learning approach to formation of earthquake categories using hierarchies of magnitude and consequence to guide emergency management,” Data Science and Management, vol. 6, no. 4, pp. 208–213, Dec. 2023, doi: https://doi.org/10.1016/j.dsm.2023.06.005
M. A. Islam, S. I. Rashid, N. U. I. Hossain, R. Fleming, and A. Sokolov, “An integrated convolutional neural network and sorting algorithm for image classification and efficient flood disaster management,” Decision Analytics Journal, vol. 7, Jun. 2023, doi: https://doi.org/10.1016/j.dajour.2023.100225
I. S. Jacobs and C. P. Bean, “Fine Particles, thin films and exchange anisotropy (effects of finite dimensions and interfaces on the basic properties of ferromagnets),” in Spin Arrangements and Crystal Structure, Domains, and Micromagnetics, G. T. Rado and H. Suhl, Eds., Academic Press, 1963.
A. Adeel et al., “A survey on the role of wireless sensor networks and IoT in disaster management,” in Geological Disaster Monitoring Based on Sensor Networks, T. S. Durrani, W. Wang, and S. M. Forbes, Eds., Switzerland: Springer Natural Hazards, 2019.
D. E. Alexander, “Social media in disaster risk reduction and crisis management,” Science and Engineering Ethics, vol. 20, pp. 717–733, 2014, doi: https://doi.org/10.1007/s11948-013-9502-z
