Fire Detection Algorithms Using Multimodal ... - Bilkent University
Fire Detection Algorithms Using Multimodal ... - Bilkent University Fire Detection Algorithms Using Multimodal ... - Bilkent University
CHAPTER 2. FLAME DETECTION IN VISIBLE RANGE VIDEO 17Table 2.1: The mean red, green and blue channel values and variances of tenGaussian distributions modeling flame color in Fig. 2.2 (b) are listed.Distribution Red Green Blue Variance1 121.16 76.87 43.98 101.162 169.08 84.14 35.20 102.773 177.00 104.00 62.00 100.004 230.42 113.78 43.71 107.225 254.47 214.66 83.08 100.116 254.97 159.06 151.00 100.087 254.98 140.98 141.93 100.398 254.99 146.95 102.99 99.579 255.00 174.08 175.01 101.0110 255.00 217.96 176.07 100.78wherec = η(Q n |µ d,n , Σ d,n ) (2.11)A Gaussian mixture model with ten Gaussian distributions is presented inFig. 2.2 (b). In this figure, spheres centered at the mean values of Gaussianshave radii twice the corresponding standard deviations. The mean red, greenand blue values and variances of Gaussian distributions in Fig. 2.2 (b) are listedin Table 2.1.Once flame pixel process is modeled and fixed in the training phase, theRGB color vector of a pixel is checked whether the pixel lies within two standarddeviations of the centers of the Gaussians to determine its nature. In other words,if a given pixel color value is inside one of the spheres shown in Fig. 2.2 (b), thenit is assumed to be a fire colored pixel. We set a binary mask, called FireColored,which returns whether a given pixel is fire colored or not. The intersection of thismask with Blobs formed in the first step is fed into the next step as a new binarymask called Fire.
CHAPTER 2. FLAME DETECTION IN VISIBLE RANGE VIDEO 182.2.3 Temporal Wavelet AnalysisThe third step of our fire detection algorithm is to keep track of the frequencyhistory of pixels in the fire colored region and analyze the history. In order todetect flicker or oscillations in pixels due to fire in a reliable manner, the videocapture rate should be high enough to capture high-frequency flicker in flames.To capture 10 Hz flicker, the video should capture at least 20 frames per second(fps). However, in some surveillance systems, the video capture rate is below 20Hz. If the video is available at a lower capture rate, aliasing occurs but flicker dueto flames can still be observed in the video. For example, 8 Hz sinusoid appearsas 2 Hz sinusoid in a 10 fps video.Figure 2.3: A two-stage filter bank. HPF and LPF represent half-band high-passand low-pass filters, with filter coefficients {− 1, 1, − 1} and { 1, 1, 1 }, respectively.4 2 4 4 2 4This filter bank is used for wavelet analysis.Each pixel I(x, n) at location x in the image frame at time step n whichalso belongs to the binary mask Fire is fed to a two stage-filter bank as shownin Fig. 2.3. The signal Ĩn(x) is a one-dimensional signal representing the temporalvariations in color values of the pixel I(x, n) at location x in the n − thimage frame. Temporal wavelet analysis can be carried out using either the luminance(Y component) in YUV color representation or the red component inRGB color representation. In our implementation the red channel values of thepixels are used. The two-channel subband decomposition filter bank is composedof half-band high-pass and low-pass filters with filter coefficients {− 1, 1, − 1} and4 2 4
- Page 3 and 4: I certify that I have read this the
- Page 7 and 8: viiiçin de işaret işleme yöntem
- Page 9: ixThis work was supported in part b
- Page 14 and 15: LIST OF FIGURESxiv2.6 Three-state M
- Page 16 and 17: LIST OF FIGURESxvi4.4 Two moving ob
- Page 18 and 19: List of Tables1.1 Organization of t
- Page 20 and 21: Chapter 1IntroductionDynamic textur
- Page 22: CHAPTER 1. INTRODUCTION 3No fire de
- Page 27 and 28: Chapter 2Flame Detection in Visible
- Page 29 and 30: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 31 and 32: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 33 and 34: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 35: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 39 and 40: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 41 and 42: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 43 and 44: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 45 and 46: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 47 and 48: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 49 and 50: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 51 and 52: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 53 and 54: CHAPTER 2. FLAME DETECTION IN VISIB
- Page 55 and 56: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 57 and 58: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 59 and 60: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 61 and 62: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 63 and 64: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 65 and 66: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 67 and 68: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 69 and 70: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 71 and 72: CHAPTER 3. FLAME DETECTION IN INFRA
- Page 73 and 74: Chapter 4Short Range Smoke Detectio
- Page 75 and 76: CHAPTER 4. SHORT RANGE SMOKE DETECT
- Page 77 and 78: CHAPTER 4. SHORT RANGE SMOKE DETECT
- Page 79 and 80: CHAPTER 4. SHORT RANGE SMOKE DETECT
- Page 81 and 82: CHAPTER 4. SHORT RANGE SMOKE DETECT
- Page 83 and 84: CHAPTER 4. SHORT RANGE SMOKE DETECT
- Page 85 and 86: CHAPTER 4. SHORT RANGE SMOKE DETECT
CHAPTER 2. FLAME DETECTION IN VISIBLE RANGE VIDEO 17Table 2.1: The mean red, green and blue channel values and variances of tenGaussian distributions modeling flame color in Fig. 2.2 (b) are listed.Distribution Red Green Blue Variance1 121.16 76.87 43.98 101.162 169.08 84.14 35.20 102.773 177.00 104.00 62.00 100.004 230.42 113.78 43.71 107.225 254.47 214.66 83.08 100.116 254.97 159.06 151.00 100.087 254.98 140.98 141.93 100.398 254.99 146.95 102.99 99.579 255.00 174.08 175.01 101.0110 255.00 217.96 176.07 100.78wherec = η(Q n |µ d,n , Σ d,n ) (2.11)A Gaussian mixture model with ten Gaussian distributions is presented inFig. 2.2 (b). In this figure, spheres centered at the mean values of Gaussianshave radii twice the corresponding standard deviations. The mean red, greenand blue values and variances of Gaussian distributions in Fig. 2.2 (b) are listedin Table 2.1.Once flame pixel process is modeled and fixed in the training phase, theRGB color vector of a pixel is checked whether the pixel lies within two standarddeviations of the centers of the Gaussians to determine its nature. In other words,if a given pixel color value is inside one of the spheres shown in Fig. 2.2 (b), thenit is assumed to be a fire colored pixel. We set a binary mask, called <strong>Fire</strong>Colored,which returns whether a given pixel is fire colored or not. The intersection of thismask with Blobs formed in the first step is fed into the next step as a new binarymask called <strong>Fire</strong>.