Title: IoT-based early forest fire detection using MLP and AROC method

Authors: V. Vinodhini; M.R. Sundara Kumar; S. Sankar; Digvijay Pandey; Binay Kumar Pandey; Vinay Kumar Nassa

Addresses: Department of CSE, Sona College of Technology, Salem – 636 005, India ' Department of CSE, Sona College of Technology, Salem – 636 005, India ' Department of CSE, Sona College of Technology, Salem – 636 005, India ' Department of Technical Education, IET, Dr. A.P.J. Abdul Kalam Technical University, Uttar Pradesh, India ' Department of IT, GovindBallabh Pant University of Agriculture and Technology, UK, India ' Department of Computer Science Engineering, South Point Group of Institutions, Sonepat – 131001, India

Abstract: The forest is a natural ecosystem that must be protected against natural calamities. Forest fire is one such calamity, and the goal of this work is to alert the event of disaster so that natural resources can be saved. The existing methods have few limitations like false alert, no timely notification, lack of network coverage, etc. The proposed work uses multi-layer perceptron (MLP) and advanced relative operating characteristic (AROC) approaches to address these constraints. The proposed model has accuracy of 90%, which is higher than the fuzzy logic and average consensus algorithm.

Keywords: forest fire; internet of things; IoT; artificial neural networks; ANNs; flame sensor; smoke sensor; multi-layer perceptron; MLP.

DOI: 10.1504/IJGW.2022.122794

International Journal of Global Warming, 2022 Vol.27 No.1, pp.55 - 70

Received: 27 Jun 2021
Accepted: 24 Sep 2021
Published online: 11 May 2022 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article