Fire Detection Algorithms Using Multimodal ... - Bilkent University

Fire Detection Algorithms Using Multimodal ... - Bilkent University Fire Detection Algorithms Using Multimodal ... - Bilkent University

theses.eurasip.org
from theses.eurasip.org More from this publisher
11.07.2015 Views

CHAPTER 5. FLAME DETECTION USING PIR SENSORS 69based system or a regular IR sensor based system is in the order of one thousanddollars.In this chapter, wavelet domain signal processing is also used, which providesrobustness against sensor signal drift due to temperature variations in the observedarea. Regular temperature changes due to hot plates and radiators areslow variations compared to the moving objects and flames. Since wavelet subsignalsof a waveform are high-pass and band-pass in nature they do not getaffected by the slow variations.Events are classified into two different classes in this approach. The firstclass represents fire events, on the other hand, the second class represents nonfireevents. PIR sensor circuits are designed for detecting the movement of hotobjects. Therefore, we include regular human motion events such as walking orrunning in the non-fire event class.The PIR sensor can be considered as a single-pixel camera without loss ofgenerality. Therefore, the proposed PIR based fire detection algorithm is obtainedsimply by removing the spatial analysis steps of the video flame detection methodsdeveloped in Chapters 2 and 3.Data acquisition and the PIR systems are described in the next Section. Theproposed algorithm and the experiments are presented in Sections 5.2 and 5.3,respectively.5.1 PIR Sensor System and Data AcquisitionCommercially available PIR sensor read-out circuits produce binary outputs.However, it is possible to capture a continuous time analog signal indicating thestrength of the received signal in time. The corresponding circuit for capturingan analog signal output is shown in Fig. 5.1.The circuit consists of 4 operational amplifiers (op amps), IC1A, IC1B, IC1Cand IC1D. IC1A and B constitute a two stage amplifier circuit whereas IC1C

CHAPTER 5. FLAME DETECTION USING PIR SENSORS 70Figure 5.1: The circuit diagram for capturing an analog signal output from a PIRsensor.and D couple behaves as a comparator. The very-low amplitude raw output atthe 2 nd pin of the PIR sensor is amplified through the two stage amplifier circuit.The amplified signal at the output of IC1B is fed into the comparator structurewhich outputs a binary signal, either 0 V or 5 V. Instead of using binary outputin the original version of the PIR sensor read-out circuit, we directly measure theanalog output signal at the output of the 2 nd op amp, IC1B.In order to capture the flame flicker process the analog signal is sampled witha sampling frequency of f s = 50Hz because the highest flame flicker frequency is13Hz [1] and f s = 50Hz is well above 2 × 13Hz. In Fig. 5.2, a frequency distributionplot corresponding to a flickering flame of an uncontrolled fire is shown. Itis clear that the sampling frequency of 50Hz is sufficient. Typical sampled signalfor no activity case using 8 bit quantization is shown in Fig. 5.3. Other typicalreceived signals from a moving person and flickering fire are presented in Fig. 5.4.The strength of the received signal from a PIR sensor increases when thereis motion due to a hot body within its viewing range. In fact, this is due tothe fact that pyroelectric sensors give an electric response to a rate of change

CHAPTER 5. FLAME DETECTION USING PIR SENSORS 70Figure 5.1: The circuit diagram for capturing an analog signal output from a PIRsensor.and D couple behaves as a comparator. The very-low amplitude raw output atthe 2 nd pin of the PIR sensor is amplified through the two stage amplifier circuit.The amplified signal at the output of IC1B is fed into the comparator structurewhich outputs a binary signal, either 0 V or 5 V. Instead of using binary outputin the original version of the PIR sensor read-out circuit, we directly measure theanalog output signal at the output of the 2 nd op amp, IC1B.In order to capture the flame flicker process the analog signal is sampled witha sampling frequency of f s = 50Hz because the highest flame flicker frequency is13Hz [1] and f s = 50Hz is well above 2 × 13Hz. In Fig. 5.2, a frequency distributionplot corresponding to a flickering flame of an uncontrolled fire is shown. Itis clear that the sampling frequency of 50Hz is sufficient. Typical sampled signalfor no activity case using 8 bit quantization is shown in Fig. 5.3. Other typicalreceived signals from a moving person and flickering fire are presented in Fig. 5.4.The strength of the received signal from a PIR sensor increases when thereis motion due to a hot body within its viewing range. In fact, this is due tothe fact that pyroelectric sensors give an electric response to a rate of change

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!