Fire Detection Algorithms Using Multimodal ... - Bilkent University
Fire Detection Algorithms Using Multimodal ... - Bilkent University Fire Detection Algorithms Using Multimodal ... - Bilkent University
CHAPTER 6. WILDFIRE DETECTION 85each incoming sample x. If the number is positive (negative), then the individualalgorithm decides that there is (not) smoke due to forest fire in the viewing rangeof the camera. Output values of decision functions express the confidence levelof each sub-algorithm. Higher the value, the more confident the algorithm.6.2.1 Detection of Slow Moving ObjectsVideo objects at far distances to the camera seem to move slower (px/sec) incomparison to the nearby objects moving at the same speed. Assuming thecamera is fixed, two background images, B fast (x, n) and B slow (x, n) correspondingto the scene with different update rates are estimated [9], [65], where x is thelocation of the pixel at frame number n.In [19] a background image B(x, n + 1) at time instant n + 1 is recursivelyestimated from the image frame I(x, n) and the background image B(x, n) of thevideo as follows:B(x, n + 1) =B(x, n){aB(x, n) + (1 − a)I(x, n) if x is stationaryif x is a moving pixel(6.1)where I(x, n) represent the intensity value of the pixel at location x in the n thvideo frame I, and a is a parameter between 0 and 1. Initially, B fast (x, 0) andB slow (x, 0) can be taken as I(x, 0). Stationary and moving pixel definitions aregiven in [19]. Background images B fast (x, n) and B slow (x, n) are updated as inEq. 6.1 with different update rates. In our implementation, B fast (x, n) is updatedat every frame and B slow (x, n) is updated once in a second with a = 0.7 and 0.9,respectively.Slow moving objects within the viewing range of the camera are detected bycomparing background images, B fast and B slow [83], [9], [65]. If there exists asubstantial difference between the two images for some period of time, then analarm for slow moving region is raised, and the region is marked.The decision value indicating the confidence level of the first sub-algorithm isdetermined by the difference between background images. The decision function
CHAPTER 6. WILDFIRE DETECTION 86D 1 (x, n) is defined as:⎧⎪⎨−1if |B fast (x, n)−B slow (x, n)|≤T lowD 1 (x, n) = 2 |Bfast (x,n)−B slow (x,n)|−T lowT high −T low−1 if T low ≤|B fast (x, n)−B slow (x, n)|≤T high⎪⎩1 if T high ≤|B fast (x, n)−B slow (x, n)|(6.2)where 0 < T low < T high are experimentally determined threshold values. In ourimplementation, T low (T high ) is taken as 10 (30) on the luminance (Y) componentof video.Confidence value is 1 (−1), if the difference |B fast (x, n)−B slow (x, n)| is higher(lower) than threshold T high (T low ).The decision function D 1 (x, n) takes realvalues in the range [-1,1] if the difference is in between the two threshold values.Smoke due to forest fires at further distances (> 5km) to the camera seemto move even slower. Therefore, smoke regions at these distances appear neitherin B fast nor B slow images. This results in lower difference values between backgroundimages B slow and B fast . In order to have substantial difference values anddetect smoke at distances further than 5km to the camera, B fast terms in Eq. 6.2are replaced by the current image I.6.2.2 Detection of Smoke-Colored RegionsWhenever a slow moving region is detected, its color content is analyzed. Smokedue to forest fires is mainly composed of carbon dioxide, water vapor, carbonmonoxide, particulate matter, hydrocarbons and other organic chemicals, nitrogenoxides, trace minerals and some other compounds [2]. The grayish color ofthe rising plume is primarily due to water vapor and carbon particles in the outputfire composition. Such regions can be identified by setting thresholds in theYUV color space. Also, luminance value of smoke regions should be high especiallyat the initial phases of a wildfire, as shown in Fig. 6.1. On the other hand,the chrominance values should be very low in a smoke region. Confidence valuecorresponding to this sub-algorithm should account for these characteristics.
- 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
- Page 87 and 88: Chapter 5Flame Detection Using PIRS
- Page 89 and 90: CHAPTER 5. FLAME DETECTION USING PI
- Page 91 and 92: CHAPTER 5. FLAME DETECTION USING PI
- Page 93 and 94: CHAPTER 5. FLAME DETECTION USING PI
- Page 95 and 96: CHAPTER 5. FLAME DETECTION USING PI
- Page 97 and 98: CHAPTER 5. FLAME DETECTION USING PI
- Page 99 and 100: CHAPTER 5. FLAME DETECTION USING PI
- Page 101 and 102: CHAPTER 6. WILDFIRE DETECTION 82In
- Page 103: CHAPTER 6. WILDFIRE DETECTION 84are
- Page 107 and 108: CHAPTER 6. WILDFIRE DETECTION 88tha
- Page 109 and 110: CHAPTER 6. WILDFIRE DETECTION 90The
- Page 111 and 112: CHAPTER 6. WILDFIRE DETECTION 92D 1
- Page 113 and 114: CHAPTER 6. WILDFIRE DETECTION 94as
- Page 115 and 116: CHAPTER 6. WILDFIRE DETECTION 96Sol
- Page 117 and 118: CHAPTER 6. WILDFIRE DETECTION 98as
- Page 119 and 120: CHAPTER 6. WILDFIRE DETECTION 100Un
- Page 121 and 122: CHAPTER 6. WILDFIRE DETECTION 102is
- Page 123 and 124: CHAPTER 6. WILDFIRE DETECTION 104Ta
- Page 125 and 126: Chapter 7Conclusion and Future Work
- Page 127 and 128: CHAPTER 7. CONCLUSION AND FUTURE WO
- Page 129 and 130: Bibliography[1] B. Albers and A. Ag
- Page 131 and 132: BIBLIOGRAPHY 112[18] D. Chetverikov
- Page 133 and 134: BIBLIOGRAPHY 114[38] G. Healey, D.
- Page 135 and 136: BIBLIOGRAPHY 116[56] U. Niesen, D.
- Page 137 and 138: BIBLIOGRAPHY 118[77] C. Stauffer an
CHAPTER 6. WILDFIRE DETECTION 86D 1 (x, n) is defined as:⎧⎪⎨−1if |B fast (x, n)−B slow (x, n)|≤T lowD 1 (x, n) = 2 |Bfast (x,n)−B slow (x,n)|−T lowT high −T low−1 if T low ≤|B fast (x, n)−B slow (x, n)|≤T high⎪⎩1 if T high ≤|B fast (x, n)−B slow (x, n)|(6.2)where 0 < T low < T high are experimentally determined threshold values. In ourimplementation, T low (T high ) is taken as 10 (30) on the luminance (Y) componentof video.Confidence value is 1 (−1), if the difference |B fast (x, n)−B slow (x, n)| is higher(lower) than threshold T high (T low ).The decision function D 1 (x, n) takes realvalues in the range [-1,1] if the difference is in between the two threshold values.Smoke due to forest fires at further distances (> 5km) to the camera seemto move even slower. Therefore, smoke regions at these distances appear neitherin B fast nor B slow images. This results in lower difference values between backgroundimages B slow and B fast . In order to have substantial difference values anddetect smoke at distances further than 5km to the camera, B fast terms in Eq. 6.2are replaced by the current image I.6.2.2 <strong>Detection</strong> of Smoke-Colored RegionsWhenever a slow moving region is detected, its color content is analyzed. Smokedue to forest fires is mainly composed of carbon dioxide, water vapor, carbonmonoxide, particulate matter, hydrocarbons and other organic chemicals, nitrogenoxides, trace minerals and some other compounds [2]. The grayish color ofthe rising plume is primarily due to water vapor and carbon particles in the outputfire composition. Such regions can be identified by setting thresholds in theYUV color space. Also, luminance value of smoke regions should be high especiallyat the initial phases of a wildfire, as shown in Fig. 6.1. On the other hand,the chrominance values should be very low in a smoke region. Confidence valuecorresponding to this sub-algorithm should account for these characteristics.