Introduction of System
Fire detection is an important issue which is related to human life and social security. Image processing based fire detector does very fast detection, as human eyes do, but it depending on camera quality if the camera is high definition then detection will be reliable. I am going to propose image processing based “Fire Detection and Extinguisher System” using Arduino.
Today all the organizations, hospitals and other companies are using different technology for fire detection. Some systems are based on different heat sensors. Some organizations like Shifa International Hospital Islamabad, MCB Bank and others banks. These organizations are using only sensor based systems for fire detection.
Limitation of Existing System
There are a lot of existing systems already available in different organizations but all of them are using only sensors. All the systems detect the fire through heat sensors. Heat sensors depends on environment if the environment become hot then sensor will detect the fire otherwise not. However, sensors are not smart thing to do this job because of its speed, and when speed is slower than the alert generation will also be slow. It also depends on distance between fire area and sensor location. Earlier systems were based only on sensor.
As introduced earlier I am going to proposed “Fire detection and Extinguisher System” for early fire detection. Early stage of fire detection and warning is imperative. However, smoke and temperature detection is slower than light detection, in order to facilitate earlier detection of fire I am introducing a new system in which I am using camera for capturing images of fire. I will train my system for recognizing fire, then that capture images are taken for color recognition and when the fire is detected then computer send signal to Arduino and Arduino send the signal to LED for alert.
Modules of Proposed System
1) Fire detection
In this module, fire detects using image processing. In this project I am using open CV and python for fire detection. I created a HAAR Cascade Classifier for fire detection using Open CV. HAAR Cascade Classifier is a trainer and detector for train our own cascade classifier, HAAR Cascade is used to detect object for which it has been trained. Lots of positive and negative images samples are need to train this HAAR Cascade Classifier. Training of cascade classifier is complex and time consuming process, so to make it easy I find a cascade training software on web name is “cascade trainer GUI”.
1) Fire Alerts
In this module when fire is detected by image processing then the system will give signal to Arduino to start alarm through buzzer.
Features of Proposed System
Early Fire Detection
The earlier a fire is detected, the faster it will be that firefighters will respond. This can mean you may avoid major damage or even worse, the complete destruction of the home or building.
A fire detection and alert alarm system gives the homeowner or building owner protection 24 hours a day, every day of the week.
Easy and Affordable
In less price you can buy this system
Tools & Languages for Development
- IDE (Pycharm)
- Programming Language (Python)
- Libraries (OpenCV, Numpy , Scipy , Pyserial)
- Arduino uno
- 16×2 LCD
- 5volt buzzer
- 5volt relay
- Bc547 transistor
- Resistors 470r, 1k, 220r, 10k preset
- Capacitors 1000uf/25volt, 470uf/16 volt
- Diode 1N4007
- Arduino cable
- Mini submersible pump