© APR 2020 | IRE Journals | Volume 3 Issue 10 | ISSN: 2456-8880

Fire Detection and Quenching System Using Image

Processing
V S MAHIMA CHOWDARY MOVVA1, THUMMAPUDI SRAVANI2, REDDYBOINA PRIYANKA3,
VEMURI SUJANI4, TADI SHALINI5
1, 2, 3, 4
Student, Department of Electronics and Communication, Vasireddy Venkatadri Institute of
Technology, Andhra Pradesh, India
5
Assistant Professor, Department of Electronics and Communication, Vasireddy Venkatadri Institute of
Technology, Andhra Pradesh, India

Abstract- Fire detection is an important issue which produce 3D point clouds from stereo cameras, stitch
is related to human life and social security. This is an images together to produce a high resolution image of
image processing based fire detection and quenching an entire scene, find similar images from an image
system using OpenCV. The other important database, remove red eyes from images taken using
components used in this project are Arduino, flash, follow eye movements, recognize scenery and
webcam, buzzer and pump. Arduino is used as establish markers to overlay it with augmented reality,
controller to control relay, buzzer and LED. The etc.
pump is connected to the relay. The system is divided
into two parts, namely Fire detection and Fire OpenCV has C++, Python, Java and MATLAB
extinguisher. In the first part, the fire is detected interfaces and it supports Windows,
using image processing. OpenCV and python are Linux, Android and Mac OS. It leans mostly towards
used for fire detection. Cascade classifier is trained real-time applications and takes advantage of MMX
and used in the detection of fire. In the second part, and SSE instructions when available. A full featured
camera preview will be shown, if fire is present, it gets CUDA and OpenCL interfaces are being developed
detected and turns ON the pump and the buzzer starts right now. There are over 500 algorithms and are about
the beep sound. 10 times as many functions that compose or support
those algorithms. OpenCV is originally written in C++
Indexed Terms- Arduino, Cascade classifier, Fire and it has an interface that works seamlessly with STL
detection, OpenCV, Python. containers.

I. INTRODUCTION II. EXISTING METHODS

OpenCV is and open source computer vision and a Existing systems include Fire Detection systems
software library in machine learning. OpenCV which employ heat sensors or temperature sensors or
provides a common infrastructure for computer vision smoke sensors or a combination of these.
applications and to accelerate the use of machine
perception in the commercial products. OpenCV is a Limitation of existing systems:
BSD-licensed product and it is easy for businesses to  It takes long time for the smoke detectors to detect
utilize and modify the code. the smoke.
 Even if there is any fire, the smoke may be
The library has more than 2500 optimized algorithms, generated quite later after burning the
which including both classic and state-of-the-art surroundings.
computer vision and machine learning algorithms.  There may be damage to the surroundings by the
These algorithms can detect and recognize faces, time fire is detected.
identify objects, classify human actions in videos,
track moving objects, extract 3D models of objects,

IRE 1702116 ICONIC RESEARCH AND ENGINEERING JOURNALS 10

 © APR 2020 | IRE Journals | Volume 3 Issue 10 | ISSN: 2456-8880

III. PROPOSED SYSTEM

The main purpose of the proposed system is to reduce

the loss of life and money during the fire accidents by
detecting and extinguishing the fire as early as
possible.

The system is basically divided into two parts:

 Fire Detection
 Fire Extinguisher

1) Fire Detection
Fig. 1: Arduino Uno module
Webcam is used for capturing the video and then the
extracted frames are checked for the fire. Cascade
3.2 Relay Module
classifier is trained to detect the fire and by using
OpenCV along with the classifier, the fire can be
The relay used is of 5v of voltage and 10A of current
detected.
with 2-channel interface board. With large current, it
can control various appliances. It can be controlled
2) Fire Extinguisher
directly with 3.3v or 5v signals from Arduino.
When fire is detected, the Arduino will turns ON the
LED, buzzer and relay which in turn will ON the
pump. There is 1*4 pin header for connecting power and for
controlling two relays.
3.1 Arduino Description

Arduino is an open source computer hardware and

software single-board microcontroller meant to make
the applications more accessible which are interactive
objects and its surroundings. It senses the environment
by receiving inputs from sensors and effects the
surroundings by controlling the actuators.

It has 14 digital input/output pins in which 6 can be

used as PWM outputs and a 16 MHz ceramic
resonator, an ICSP header, a USB connection, 6
analog inputs, a power jack and a reset button.
Fig2: Relay module
Features of Arduino:
3.3 Buzzer
 Microcontroller : ATmeg328
 Operating voltage : 5v
A buzzer or beeper is an audio signaling device, which
 Input voltage : 7v – 12v
may be mechanical, electromechanical, or
 Flash Memory : 32KB
piezoelectric. The typical uses of buzzers and beepers
 SRAM : 2KB include alarm devices, timers, and confirmation of
 Clock Frequency : 16MHZ user input such as a mouse click or keystroke.
 Digital I/O pins : 14
 Analog input pins : 6

IRE 1702116 ICONIC RESEARCH AND ENGINEERING JOURNALS 11

 © APR 2020 | IRE Journals | Volume 3 Issue 10 | ISSN: 2456-8880

The below figure depicts the output of the project

when the fire is not detected

Fig3: Buzzer

3.4 Operation

The sequence of steps for circuit operation are:

1. When the power is given to Arduino through
USB, initially the relay is OFF.
When the fire is not detected, the pump is OFF
2. The webcam will be continuously capturing the
video, and frames are extracted from it.
3. Each frame is checked for the presence of fire. It
is done by using the HAAR cascade classifier.
4. If the fire is detected, the Arduino will turn ON
the relay, LED and the buzzer.
5. When the relay is ON, it will turn ON the pump.
6. The pump will be ON until the fire is quenched or
extinguished, which means none of the frames has
fire.

IV. RESULTS
The below figures depict the output of the project
In industries, if there is fire, it may cause a lot of
when there is fire, the screen shows the message fire is
damage and there would be a huge loss of money. So
detected.
the sole purpose of this project is to detect the fire as
early as possible and to extinguish it. So that there will
not be any damage to the equipment and no loss of
money and life. The CCTVs installed previously in the
industries can be used for extracting images so the cost
required will be less.

IRE 1702116 ICONIC RESEARCH AND ENGINEERING JOURNALS 12

 © APR 2020 | IRE Journals | Volume 3 Issue 10 | ISSN: 2456-8880

When fire is detected, the pump is turned ON [7] MP Sivaram kumar; S Rajasekaran; “Detection
and extinguishing forest fires using wireless sensor
and actor networks”

CONCLUSION

The fire detection and quenching system can reduce

the loss of life and money when there is an occurrence
of fire. It does so by detecting and extinguishing the
fire when it is still small. It reduces the time required
for locating the fire and it provides the service to the
owner 24 hours a day. It is very cheap since the CCTV
can be used for extracting the frames and the hardware
required is also minimal.

Future studies will focus on determining the exact

point of occurrence of fire and providing the
appropriate evacuation guidance to the people in that
area.

REFERENCES

[1] C Yu, X Zhang-“A real time video fire flame and

smoke detection algorithm”
[2] www.electronicshub.org
[3] https://randomnerdtutorials.com/guide-for-relay-
module-with-arduino
[4] High-rise building fire safety standards
development research, Korea fire safety
Association(2012)
[5] P.G Smith- Fire Detection and alarm systems,
ieee.org volume 1977, issue 3
[6] Huide Liu; Lili Gao; Suwei Li; Tao Wu “About
automatic fire alarm systems research”, The 2 nd
IEEE International Conference on Information
Management and Engineering (ICIME).

IRE 1702116 ICONIC RESEARCH AND ENGINEERING JOURNALS 13