Skip to main content

ORIGINAL RESEARCH article

Front. Environ. Sci.
Sec. Big Data, AI, and the Environment
Volume 12 - 2024 | doi: 10.3389/fenvs.2024.1440396
This article is part of the Research Topic Artificial Intelligence in Environmental Engineering and Ecology: Towards Smart and Sustainable Cities View all 7 articles

Deep Learning with Ensemble Approach for Early Pile Fire Detection Using Aerial Images

Provisionally accepted
Dhyey Divyeshkumar Joshi Dhyey Divyeshkumar Joshi 1,2Satish Kumar Satish Kumar 1,3*Shruti Patil Shruti Patil 1,3Pooja Kamat Pooja Kamat 3Shrikrishna Kolhar Shrikrishna Kolhar 3Ketan Kotecha Ketan Kotecha 1,3,4*
  • 1 Symbiosis Centre of Applied Artificial Intelligence, Symbiosis International University, Pune, India
  • 2 Devang Patel Institute of Advance Technology and Research, Charotar University of Science and Technology, Charusat, India
  • 3 Symbiosis Institute of Technology, Symbiosis International University, Pune, India
  • 4 Peoples' Friendship University of Russia, Moscow, Russia

The final, formatted version of the article will be published soon.

    Wildfires rank among the world's most devastating and expensive natural disasters, destroying vast forest resources and endangering lives. Traditional firefighting methods, reliant on ground crew inspections, have notable limitations and pose significant risks to firefighters. Consequently, dronebased aerial imaging technologies have emerged as a highly sought-after solution for combating wildfires. Recently, there has been growing research interest in autonomous wildfire detection using drone-captured images and deep-learning algorithms. This paper introduces a novel deep-learningbased method, distinct in its integration of infrared thermal, white, and night vision imaging to enhance early pile fire detection, thereby addressing the limitations of existing methods. The study evaluates the performance of machine learning algorithms such as random forest (RF) and support vector machines (SVM), alongside pre-trained deep learning models including AlexNet, Inception ResNetV2, InceptionV3, VGG16, and ResNet50V2 on thermal-hot, green-hot, and white-green-hot color images. The proposed approach, particularly the ensemble of ResNet50V2 and InceptionV3 models, achieved over 97% accuracy and over 99% precision in early pile fire detection on the FLAME dataset. Among the tested models, ResNet50V2 excelled with the thermal-fusion palette, InceptionV3 with the white-hot and green-hot fusion palettes, and VGG16 with a voting classifier on the normal spectrum palette dataset. Future work aims to enhance the detection and localization of pile fires to aid firefighters in rescue operations.

    Keywords: aerial imaging1, deep learning2, fire piles monitoring3, thermal infrared images4, RGB images5, wildfire detection system6

    Received: 29 May 2024; Accepted: 06 Sep 2024.

    Copyright: © 2024 Divyeshkumar Joshi, Kumar, Patil, Kamat, Kolhar and Kotecha. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

    * Correspondence:
    Satish Kumar, Symbiosis Centre of Applied Artificial Intelligence, Symbiosis International University, Pune, India
    Ketan Kotecha, Symbiosis Centre of Applied Artificial Intelligence, Symbiosis International University, Pune, India

    Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.