Fired:An Image-Based Fire Detector for Smart Homes using Machine Learning Algorithms

Smart homes are to be protected from fire hazards which is a crucial safety concern. Existing ways of detecting fire is time consuming, hence causing maximum injuries and financial loss, so we have come up with an effective solution for detecting fire in smart homes using image processing and Machin...

Full description

Saved in:

Bibliographic Details
Published in:	2024 International Conference on Knowledge Engineering and Communication Systems (ICKECS) Vol. 1; pp. 1 - 4
Main Authors:	Srinath, G M, Yashaswini, K M, Sahana Shree, N, Supraja, B, Varshitha, V, Arfa Thareen, K
Format:	Conference Proceeding
Language:	English
Published:	IEEE 18-04-2024
Subjects:	CNN(Convolutional Neural Networks) Containers Docker FireD (Fire Device) Image processing Machine learning Machine learning algorithms Neural networks Platform as a service Smart homes YOLO Yolov5 and Yolov8(You Only Look Once)
Online Access:	Get full text
Tags:	Add Tag No Tags, Be the first to tag this record!

Description
Summary:	Smart homes are to be protected from fire hazards which is a crucial safety concern. Existing ways of detecting fire is time consuming, hence causing maximum injuries and financial loss, so we have come up with an effective solution for detecting fire in smart homes using image processing and Machine Learning techniques. FireD extracts the features from photos and captured video frames using Convolutional Neural Network (CNN) and in turn sends them to Neural Network which implements clustering algorithm known as Yolov5 to ensure the ability to classify images as Fire and Non-Fire. Once the fire is detected, model will send the images to Heroku cloud which serves as container based cloud Platform as a Service (PaaS) to access, deploy and manage the images. On the detection of fire Notification is sent to Telegram through telegram bot. Overall performance data from model indicates that FireD is likely to perform well in real-world scenarios. Therefore, FireD is more reliable than other existing systems since FireD application is pushed to the docker container with all the required libraries and files. The application can be accessed and implemented easily.
DOI:	10.1109/ICKECS61492.2024.10616511