0% found this document useful (0 votes)
6 views

Fire Detection With Image Processing

: Convolutional neural networks (CNNs) have yielded state-of-theart performance in image classification and other computer vision tasks. Their application in fire detection systems will substantially improve detection accuracy, which will eventually minimize fire disasters and reduce the ecological and social ramifications. However, the major concern with CNNbased fire detection systems is their implementation in realworld surveillance networks, due to their high memory and computational require
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
6 views

Fire Detection With Image Processing

: Convolutional neural networks (CNNs) have yielded state-of-theart performance in image classification and other computer vision tasks. Their application in fire detection systems will substantially improve detection accuracy, which will eventually minimize fire disasters and reduce the ecological and social ramifications. However, the major concern with CNNbased fire detection systems is their implementation in realworld surveillance networks, due to their high memory and computational require
Copyright
© © All Rights Reserved
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 8

11 V May 2023

https://doi.org/10.22214/ijraset.2023.52073
International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 11 Issue V May 2023- Available at www.ijraset.com

Fire Detection with Image Processing


Prof. Shruti P Naik1, Vishal Lohbande2, Shreyas Hambir3, Rohit Korade4, Rahul Hatkar5
1
Assistant Professor, Department of Computer Engineering, 2BE in Computer Engineering, JSPM NTC, Pune, Maharashtra, India

Abstract: Convolutional neural networks (CNNs) have yielded state-of-theart performance in image classification and other
computer vision tasks. Their application in fire detection systems will substantially improve detection accuracy, which will
eventually minimize fire disasters and reduce the ecological and social ramifications. However, the major concern with CNN-
based fire detection systems is their implementation in realworld surveillance networks, due to their high memory and
computational requirements for inference. In this paper, we propose an original, energy-friendly, and computationally efficient
CNN architecture, inspired by the SqueezeNet architecture for fire detection, localization, and semantic understanding of the
scene of the fire. It uses smaller convolutional kernels and contains no dense, fully connected layers, which helps keep the
computational requirements to a minimum. Despite its low computational needs, the experimental results demonstrate that our
proposed solution achieves accuracies that are comparable to other, more complex models, mainly due to its increased depth.
Moreover, this paper shows how a tradeoff can be reached between fire detection accuracy and efficiency, by considering the
specific characteristics of the problem of interest and the variety of fire data
Keywords: Fire detection, CNN, Machine learning.

I. INTRODUCTION
Rate of forest fires reports have increased yearly due to human causes and dry climate. To avoid terrible disaster of fire, many
detection techniques have been widely studied to apply in practice. Most of traditional method are based on sensors due to its low-
cost and simple installation [1]–[3]. These systems are not applicable for using outdoor where energy of flame affected by fire
materials and the burning process affected by environment that have potential cause of false alarms. Visual-based approach of
image or video processing was shown to be more reliable method to detect the fire since the closed circuit television (CCTV)
surveillance systems are now available at many public places, can help capture the fire scenes. In order to detect fire from scenes of
colour-videos, various schemes have been studied, mainly focus on the combination of static and dynamic characteristics of fire
such as colour information, texture and motion orientation, etc. Colour-based detection methods mainly depend on chosen value of
thresholds resulted in high false alarm rate; that need to be improved by extracting dynamic features of fire from sequence of images
captured in video. However, those systems are still not practical to use in large scale and hard-to-reach regions like remote and wild
forests, where the configuration and maintenance of the system are difficult tasks

II. LITERATURE SURVEY


Name Description
Year
2019 Research on In order to detect and alarm early fire timely and effectively, traditional temperature and
Image Fire smoke fire detectors are vulnerable to environmental factors such as the height of monitoring
Detection space, air velocity, dust. An image fire detection algorithm based on support vector machine
Based on is proposed by studying the features of fire in digital image.
Support
Vector
Machine
Author:Ke
Chen,Yanying
Cheng.

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 2123
International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 11 Issue V May 2023- Available at www.ijraset.com

2018 Using Popular In this paper, we focus on three problems that surrounded forest fire detection, real-time,
Object early fire detection, and false detection. For the first time, we use classical objective
Detection detection methods to detect forest fire: Faster R-CNN, YOLO (tiny-yolo-voc, tiny-yolo-
Methods for voc1, yolo-voc.2.0, and yolov3), and SSD, among them SSD has better real-time property,
Real Time higher detection accuracy and early fire detection ability.
Forest Fire
Detection

Year Name Description


2018 Automatic Fire and Smoke The technologies underlying fire and smoke detection systems play a
Detection Method for crucial role in ensuring and delivering optimal performance in modern
Surveillance Systems Based surveillance environments. In fact, fire can cause significant damage to
on Dilated CNNs lives and properties. Considering that the majority of cities have
Author name: Yakhyokhuja already installed camera-monitoring systems, this encouraged us to
Valikhujaev 1 , Akmalbek take advantage of the availability of these systems to develop cost-
Abdusalomov 1 effective vision detection methods.
2018 Fire Smoke Detection Based Smoke detection based on automatic visual system has been applied to
on Contextual Detection fire alarm in open spaces where traditional smoke detection system is
Author: Xuan Zhao,Hang Ji not suitable for it. However, detecting the course of smoke posed great
challenges for both systems. To address this problem, we propose a
new method that combines contextaware framework with automatic
visual smoke detection.

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 2124
International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 11 Issue V May 2023- Available at www.ijraset.com

III. METHODOLOGY
1) Preprocessing: The aim of pre-processing is an improvement of the image data that suppresses unwilling distortions or
enhances some image features important for further processing, although geometric transformations of images (e.g. rotation,
scaling, translation) are classified among pre-processing methods here since similar image resizing, converting images to
grayscale, and image augmentation.
2) Image Detection: This section covers the detail of the proposed fire pixel classification algorithm. Figure shows the flow chart
of the proposed algorithm. Rule based colour model approach has been followed due to its simplicity and effectiveness. For
that, colour space RGB and YCbCr is chosen. For classification of a pixel to be fire we have identified seven rules. If a pixel
satisfies these seven rules, we say that pixel belong to fire class. 2) Feature Extraction: Feature extraction refers to the process
of transforming raw data into numerical features that can be processed while preserving the information in the original data set.
It yields better results than applying machine learning directly to the raw data.
3) Fire detection: We took two sequential images from video frames. After applying basic two methods edge detection and colour
detection we get probable area of fire pixel then we compare the RGB value to of frame1 to the frame 2 for corresponding pixel
and if pixel value differs then motion detector will show motion and will give resultant output to the operator.
4) Flow of execution: When the fire is detect to the module, those image is gives to module as a input and when according to the
input the by the use of multithreading one thread is gives to output with massage to Owner “Fire Detected……Fire Detected
and beep signal with sound” and other and to Fire Extinguish Depart ment with Address “Emergency….Emergency and
Address (Shanti Niwas near JSPM College, Narhe, Pune)”.

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 2125
International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 11 Issue V May 2023- Available at www.ijraset.com

IV. SYSTEM ARCHITECTURE

Fig. System Architecture

V. SYSTEM IMPLEMENTATION
A. Convolutional Neural Network(CNN)
1) In deep learning a convolutional neural network (CNN) is a class of deep neural networks, most commonly applied to analyze
visual imagery. Now when we think of a neural network we think about matrix multiplications but that is not the case with
ConvNet. It uses a special technique called Convolution.
2) It has four steps:
Convo 2D
Max Pooling
Flatten
Fully connected Network
CNNs are used for image classification and recognition because of its high accuracy. The CNN follows a hierarchical model which
works on building a network, like a funnel, and finally gives out a fully-connected layer where all the neurons are connected to each
other and the output is processed

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 2126
International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 11 Issue V May 2023- Available at www.ijraset.com

OutPut

VI. CONCLUSION
In summary, an aerial based forest fire detection method has been examined through a large database of videos of forest fires of
various scene conditions. To enhance the detection rate, at first the chromatic and motion features of forest fire are extracted and
then corrected using rule to point out the fire area. Secondly, to overcome the challenge of heavy smoke that covers almost the fire,
smoke is also extracted using our proposed algorithm. Our framework proves its robustness with high accuracy rate of detection and
low false alarm rate in practical application of aerial forest fire surveillance.

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 2127
International Journal for Research in Applied Science & Engineering Technology (IJRASET)
ISSN: 2321-9653; IC Value: 45.98; SJ Impact Factor: 7.538
Volume 11 Issue V May 2023- Available at www.ijraset.com

VII. FUTURE SCOPE


The Project has been motivated by the desire a system that can detect fires and take appropriate action, without any human
intervention.
1) Implementation in a satellite to detect the accendently fire happens in the forest.
2) For further accuracy use of Neural Networks for decision making can be made and GSM module can also be implemented for
sending SMS to nearby fire station in case of severe fire. Water sprinklers can also be incorporated. By research and analysis,
the efficiency of the proposed Fire detection system can be increased. The margin of false alarms can be reduced even further
by developing algorithms to eliminate the detection of red coloured cloth as fire. By proper analysis, suitable location height
and length for camera installment can be decided, in order to remove blind-spot areas.

REFFERENCES
[1] AAAlkhatib, “Smart and Low Cost Technique for Forest Fire Detection using Wireless Sensor Network,” Int. J. Comput. Appl., vol. 81, no. 11, pp. 12–18,
2013.
[2] J. Zhang, W. Li, Z. Yin, S. Liu, and X. Guo, “Forest fire detection system based on wireless sensor network,” 2009 4th IEEE Conf. Ind. Electron. Appl. ICIEA
2009, pp. 520– 523, 2009.
[3] A. A. A. Alkhatib, “A review on forest fire detection techniques,” Int. J. Distrib. Sens. Netw., vol. 2014, no. March, 2014.
[4] P. Skorput, S. Mandzuka, and H. Vojvodic, “The use of Unmanned Aerial Vehicles for forest fire monitoring,” in 2016 International Symposium ELMAR,
2016, pp. 93–96.
[5] F. Afghah, A. Razi, J. Chakareski, and J. Ashdown, Wildfire Monitoring in Remote Areas using Autonomous Unmanned Aerial Vehicles. 2019.
[6] Hanh Dang-Ngoc and Hieu Nguyen-Trung, “Evaluation of Forest Fire Detection Model using Video captured by UAVs,” presented at the 2019 19th
International Symposium on Communications and Information Technologies (ISCIT), 2019, pp. 513–518.
[7] C. Kao and S. Chang, “An Intelligent Real-Time Fire-Detection Method Based on Video Processing,” IEEE 37th Annu. 2003 Int. Carnahan Conf. OnSecurity
Technol. 2003 Proc., 2003.
[8] C. E. Premal and S. S. Vinsley, “Image Processing Based Forest Fire Detection using YCbCr Colour Model,” Int. Conf. Circuit Power Comput. Technol.
ICCPCT, vol. 2, pp. 87–95, 2014.
[9] C. Ha, U. Hwang, G. Jeon, J. Cho, and J. Jeong, “Vision-based fire detection algorithm using optical flow,” Proc. - 2012 6th Int. Conf. Complex Intell. Softw.
Intensive Syst. CISIS 2012, pp. 526–530, 2012.
[10] K. Poobalan and S. Liew, “Fire Detection Algorithm Using Image Processing Techniques,” Proceeding 3rd Int. Conf. Artif. Intell. Comput. Sci., no. December,
pp. 12– 13, 2015

©IJRASET: All Rights are Reserved | SJ Impact Factor 7.538 | ISRA Journal Impact Factor 7.894 | 2128

You might also like