臺灣博碩士論文加值系統

為了在開放式空間中偵測火焰及煙霧，本論文統合了火焰及煙霧兩者相同的特性來偵測開放式空間中是否有火災和煙霧的存在。首先先偵測影像中是否有移動的物體，接著利用火焰本身的顏色及煙霧所造成背景影像的彩度(chrominance)的改變來對移動區域進行分析，並結合離散小波轉換(Discrete Wavelet Transformation)來分析疑似火焰及煙霧區域的邊界及兩者閃爍的程度，最後使用火焰及煙霧的邊界本身為不規則形狀的特徵來減低誤判的機會。最後將以上所有條件判別的結果相結合，來決定影像中是否有火焰及煙霧的存在。
由於用上述的方法來對煙霧及火焰進行偵測時誤判率偏高，因此本論文在最後提出了幾個方式來改善誤判率以期望提升效能，包括了改善移動偵測的門檻值決定方法、改善時間上頻率之判別方法，以及使用不同的輪廓判斷方法來判別煙霧的輪廓。

In order to detect fire and smoke in the open space, this paper integration the same properties of the flame and smoke to detect whether an open space in the presence of fire and smoke. First, detection image if there are moving objects, then use the color of flame and the chrominance value change in background image caused by smoke to analysis the move region. The third and fourth step are used the frequency to analysis the boundaries and flicker or oscillations in a pixel due to fire and smoke. The last step is using the boundaries properties of fire and smoke to reduce the false positives. Final combined the results of all steps to determine whether the video presence fire and smoke.
Because of using the above methods to detect fire and smoke will cause the false alarm rate is too high, at the end of this paper suggested several ways to improve the false alarm rate. Including improve the threshold of moving objects detection method and improve the time frequency identification methods and using the different methods to determine the contours of smoke.

摘要I
Abstract II
致謝III
目錄IV
圖目錄VI
表目錄VII
第一章 緒論1
1.1 動機1
1.2 研究目標2
1.3 論文架構3
第二章 背景知識4
2.1 移動物件偵測(Moving Object Detection)4
2.2 離散小波轉換6
第三章 研究方法與步驟9
3.1 移動偵測10
3.2 顏色及彩度分析10
3.2.1 顏色分析10
3.2.2 彩度分析11
3.3 空間上之頻率分析13
3.3.1 利用頻率分析是否為火焰13
3.3.2 利用頻率分析是否為煙霧15
3.4 時間上之頻率分析17
3.5 形狀分析21
第四章 實驗結果22
第五章 結論27
第六章 未來展望28
參考文獻29
附錄A 測試影像資料31
圖目錄
FIGURE 1 移動物件偵測的問題5
FIGURE 2 移動物件偵測之輸入影像與結果6
FIGURE 3 DWT 階層意示圖6
FIGURE 4 DWT頻帶分解圖7
FIGURE 5 DWT運算原始影像8
FIGURE 6 DWT結果9
FIGURE 7 步驟一輸出結果10
FIGURE 8 背景影像及其U跟V分量圖12
FIGURE 9 畫面中有煙霧時及其U跟V分量圖13
FIGURE 10 一般紅色區域(A)及其三個高頻頻帶係數絕對值總和(B)14
FIGURE 11 火焰區域(A)及其三個高頻頻帶係數絕對值總和(B)15
FIGURE 12 背景影像之2D-DWT結果15
FIGURE 13 影像中有煙霧出現之2D-DWT結果16
FIGURE 14 對影像頻率進行分析之方法17
FIGURE 15 時間上之頻率分析意示圖19
FIGURE 16 測試影像19
FIGURE 17 藍點時間軸上之畫素值變化(A)及頻率變化圖(B)(C)20
FIGURE 18 紅點時間軸上之畫素值變化圖(A)及頻率變化圖(B)(C)21
FIGURE 19 火焰形狀分析示意圖22
FIGURE 20 火焰偵測24
FIGURE 21 近距離煙霧偵測25
FIGURE 22 遠距離煙霧偵測25
FIGURE 23 火焰偵測之誤判25
FIGURE 24 遠距離煙霧偵測之誤判26
表目錄
TABLE 1 火焰偵測之結果23
TABLE 2 煙霧偵測之結果24

[1]Aird, B., Brown, A., 1997. Detection and alarming of the early appearance of fire using CCTV cameras. In: Nuclear Engineering Internat. Fire & Safety Conference, London, 24–26 February.
[2]B. U. Toreyin, Y. Dedeoglu, and A. E. Cetin , “Computer vision based method for real-time fire and flame detection”, Pattern Recognition Letter 27, 2006, pp.49-58.
[3]B. U. Toreyin, Y. Dedeoglu, and A. E. Cetin , “Fire detection in infrared video using wavelet analysis”, SPIE Optical Engineering, vol. 46(6), 2007.
[4]B. U. Toreyin, Y. Dedeoglu, and A. E. Cetin , “Flame Detection In Video Using Hidden Markov Models”, In ICIP’05, 2005, pp.1230-1233.
[5]B. U. Toreyin, Y. Dedeoglu, and A. E. Cetin , “Wavelet Based Real-time Smoke Detection in Video”, in EUSIPCO ’05, 2005.
[6]Cappellini V., Mattii L., Mecocci A., 1989. “An Intelligent System for Automatic Fire Detection in Forests”. University of Florence, Italy.
[7]Healey G., Slater D., Lin T., Drda B., Goedeke A.D., 1993. “ A system for real-time fire detection”. In IEEE Comput. Soc. Conf. on Computer Vision and Pattern Recognition, 15–17 June, pp. 605–606.
[8]R.T.Collins, A.J.Lipton, and T.Kanade, “A system for video surveillance and monitoring,” In 8th Int. Topical Meeting on Robotics and Remote Systems. 1999, American Nuclear Society.
[9]T. Chen, P. Wu, and Y. Chiou, “A nearly fire-detection Method based on image processing,” in ICIP’04,2004, pp.1707–1710
[10]Thou-Ho Chen, Cheng-Liang Kao, Sju-Mo Chang, 2003. “An Intelligent real-time fire detection method based on video processing”. In IEEE 37th Annual 2003 International Carnahan Conference on Security Technology, October 14-16, pp. 104-111
[11]Vicente, Jerome, Guillemant, Philippe, 2002. “An image processing technique for automatically detecting forest fire”. Internat. J. Therm. Sci. (4), 1113–1120.
[12]Wieser, Dieter, Brupbacher, Thomas, 2001. “Smoke detection in tunnels using video images”. In: 12th Internat. Conf. on Automatic Fire Detection, March 25–28, Maryland, USA.
[13]Yamagishi, H., Yamaguchi, J., 1999. “Fire flame detection algorithm using a color camera”, 1999. MHS ’99. In: Proc. 1999 Internat. Symposium on Micromechatronics and Human Science, 23–26 November, pp. 255–260.
[14]Yamagishi, H., Yamaguchi, J., 2000. A contour fluctuation data processing method for fire flame detection using a color camera. In: IEEE 26th Annual Conf. on IECON of the Industrial Electronics Society, vol. 2,22–28 October, pp. 824–829.