SMOKE DETECTION SYSTEM

Abstract

Nowadays, fire is one the most threating problems which impact peoples life. Each year in England
and Wales about 800 people die as a result of fires in buildings. A large proportion of these deaths
occur in very ordinary domestic homes. In this project, a new method was used to detect including
smoke sensor, battery, microcontroller, computer and voltage regulator. This design is very
effective both performance and cost.

Introduction
Smoke detectors have been utilized for a number of years. However, until recently, circuitry
required for these devices was rather complicated and it wasnt enough low cost consumer use. At
the same time, camera technology is available for detection of smoke recently. But, it is also
expensive than sensor technology.
In recent years many study have been done for early-fire alarm system, but most of them are
focusing on flame detection [1,2]. Smoke detectors and gas detectors have been utilized separately
for a number of years. However, until recently, circuitry required for these devices was rather
complicated and did not lend itself to low cost consumer use. As a result of increasing market
demand, integrated circuits have been developed which integrate a number of functions for smoke
alarms onto a single chip[3]. Recently, Celik et al. proposed a generic model for fire colour [4],
[5]. The authors combined their model with simple moving object detection. The objects are
identified by the background subtraction technique [6]. Recently, Toreyin et al. proposed a generic
model for fire colour [7-8]. Kopilovic et. al. in [9] propose to analyze the motion irregularities of
grayish region using the optical flow. Although these approaches are effective in most simple cases
they produce many false positives when grayish moving objects are present in the scene.


Methods/Experiments
This project is comprised of hardware and software development. The overall block diagram of the
system monitoring is shown in Figure1 that consists of gas sensor, microcontroller, battery, 5V
regulator and computer.


Figure 1

MQ2 sensor is a device that senses smoke, typically as an indicator of fire. Commercial and
residential security devices issue a signal to a fire alarm control panel as part of a fire alarm system,
while household detectors, known as smoke alarms, generally issue a local audible or visual alarm
from the detector itself.
Battery has 7.4V. So, a voltage regulator (LM7805) is used to provide 5V fix output for
microcontroller and sensor. Microcontroller gets sensor data values on analog digital channels and
it sends the values to computer with serial port communication (RS232).
An algorithm was written depending on test values. System was tested for different environment
conditions. A threshold value was determined. Flow chart of this algorithm is given figure 2.


Figure 2

Results/Discussions
A test area was selected out of the company. Test result showed that sensor gave maximum 2.2V
when there is no smoke. So, threshold value was chosen 2.7V. If microcontroller gives 2.7V and
more for 5 minutes, buzzer will be on. Otherwise, it will be off. Test results were shown in figure
3 and 4.

When there is no smoke in the company,

Figure 3
When there is smoke in the company,

Figure 4




Conclusion
Nowadays, smoke is the most important parameter to detect fire. Many studies was available. But
generally, these sytems are very expensive. Because, all of them is related with image processing.
So, a camera is increasing cost. But, system designed by me is only $30 and it gives succesful
results.

Summary
These system can be used for smart houses and forests. If we increase the numbers of sensors and
add GSM module, it can be more effective to detect for wide areas.

References
[1] B. U. Toreyn, Y. Dedeoglu and A. E. Cetin. Wavelet based real-time smoke detection in
video. EUSIPCO 05, 2005.
[2] Thou-Ho (Chao-Ho) Chen, Yen-Hui Yin, Shi-Feng Haung and Yan-Ting Ye. The smoke
detection for early fire alarming system base on video processing. Proceeding of the 2006
international Conference in intelligent information Hiding and Multimedia Signal Processing (IIH-
MSP06).

[3] R. Cucchiara, C. Grana, M. Piccardi, and A. Prati, Detecting moving objects, ghosts and
shadows in video streams, IEEE Trans. on PAMI. Vol. 25, no. 10, pp. 1337-1342, Oct. 2003.

[4] Celik, T., Demirel, H., Ozkaramanli, H., Uyguroglu, M., Fire Detection in Video Sequences
Using Statistical Color Model, Proc. Internat. Conf. on Acoustics, Speech, and Signal Processing,
vol. 2,no.pp. II-213 - II-216, May 2006.

[5] Celik, T., Demirel, H., Ozkaramanli, H.,Automatic Fire Detection in Video Sequences,
European Signal Processing Conference, EUSIPCO-06, Sept. 2006.

[6] Turgay Celik, Hasan Demirel, Huseyin Ozkaramanli, Mustafa Uyguroglu, Fire Detection
Using Statistical Color Model in Video Sequences, Journal of Visual Communication and Image
Representation (2007), doi:10.1016/j.jvcir.2006.12.003.

[7] Treyin, B.U., Dedeolu, Y., Gdkbay, U., etin, A.E, Computer vision based method for
real-time fire and flame detection, Pattern Recognition Lett., 27 (1-1), 49-58, 2006.
[8] Treyin, B.U., Dedeolu, Y., etin, A.E.,Flame detection in video using hidden Markov
models, Proc. IEEE Internat. Conf. on Image Processing, pp. 1230-1233, 2005.

[9] I. Kopilovic, B. Vagvolgyi, T. Sziranyi, Application of panoramic annular lens for motion
analysis tasks: surveillance and smoke detection, In Proceeding of 15th International Conference
on Pattern Recognition Vol. 4. IEEE, 3-7 Sept. (2000) 714-717