INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
Wildfires, or similar events in other places, are among the most common yet unfavorable occurrences brought on by climate change and rising temperatures. Thus, sophisticated yet user-friendly systems are required, ones that at the very least make it possible to employ modern tools and solutions efficiently. In order to guarantee the security and safety of diverse settings, fire and smoke detection are essential jobs. In this research, we use deep learning techniques to propose a comprehensive solution for smoke and fire detection. The project is being developed in Python, making use of the MobileNet architecture's potent capabilities. Accurately identifying fire and smoke instances in many scenarios—including pictures, videos, and real-time webcam feeds—is the primary goal of this study. The excellent accuracy shows that the model can reliably detect fire, smoke, and typical occurrences in a variety of scenarios. The suggested solution offers real-time analysis of photos, videos, and live webcam feeds in addition to multipurpose detection. This adaptability guarantees that the solution may be applied to a variety of situations, including emergency response management, fire alarm systems, and surveillance systems.
Keywords:
Convolutional Neural network, Deep Learning, Image Classification, Fire Detection, MobileNet Architecture.
Cite Article:
"Deep Learning - Based Fire and Smoke Detection System with MobileNet Architecture", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.9, Issue 4, page no.419 - 424, April-2024, Available :http://www.ijsdr.org/papers/IJSDR2404064.pdf
Downloads:
000338172
Publication Details:
Published Paper ID: IJSDR2404064
Registration ID:210732
Published In: Volume 9 Issue 4, April-2024
DOI (Digital Object Identifier):
Page No: 419 - 424
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn