臺灣博碩士論文加值系統

本研究目的在研究探討森林田野火災火焰偵測，希望能使用數位影像視頻之辨識方法，可以在將來應用在火災偵測預防系統。

目前根據內政部消防署歷年所統計，火災發生的種類以發生在建築物的件數超過總發生件數的75%以上，所以因應市場規模及人口密集程度等因素，應用在建築物防火之軟硬體系統較為成熟，廠商所開發之相關設備及系統也有較多的選項。類型大略分為利用火焰溫度之感溫型、火焰光之感光型、火焰煙霧之感煙型和火焰氣體化學成分之氣敏型……等火焰不同特性的偵測器，來對火災進行探測。

本論文聚焦於森林田野的火災偵測，考慮使用數位影像處理方法來偵測火焰發生。利用數位影像處理方法的火焰偵測方法，已經有許多學者提出相關演算法，例如支持向量機、小波分析、高斯混和模型、卡夫曼濾波、粒子群最佳化等，本研究嘗試使用數位影像處理的邊緣檢測之角點偵測方法，並配合K-mean分群演算法、區域成長法、影格差分法來提高角點偵測準確度，並去除誤判角點之數量。

The purpose of this study is to study the detection of forest fire detection in the field of fire, hoping to use digital image video recognition method, which can be used in fire detection and prevention system in the future.

At present, according to statistics from the Fire Department of the Ministry of the Interior over the years, the number of fires occurring in buildings exceeds 75% of the total number of occurrences. Therefore, due to factors such as market size and population density, System is more mature, manufacturers have developed related equipment and systems also have more options. Types are broadly divided into the flame temperature of the use of temperature-sensitive, flame-sensitive light type, flame smoke and flame gas chemical composition of the gas-type flame detector of different characteristics, to detect the fire.

This paper focuses on fire detection in the forest field, considering the use of digital image processing methods to detect flames. Many researchers have put forward the relevant algorithms, such as support vector machine, wavelet analysis, Gaussian mixture model, KFM, particle swarm optimization and so on, using the digital image processing method of flame detection. This study attempts to use digital images Corner detection method of edge detection, and K-mean clustering algorithm, regional growing method, background subtraction method to improve the accuracy of corner detection, and remove the number of false corners.

誌 謝 I
摘 要 II
ABSTRACT IV
圖目錄 VIII
表目錄 XI
第一章 緒論 1
1.1 研究背景 1
1.2 研究動機及目的 5
1.3 研究方法流程及架構 8
第二章 文獻探討 9
2-1 RGB色彩模型系統 10
2-2 HSI色彩空間系統轉換 14
2-2-1 從RGB色彩空間轉換到HSI色彩空間 17
2-2-2 從HSI色彩空間轉換到RGB色彩空間 18
2-3 YCBCR色彩空間轉換 20
2-4 火焰的色彩模型分析討論 22
2-4-1 靜態火焰的色彩分析 22
2-4-2 動態火焰的分析 25
2-5 區域成長法 28
2-6 影格差分法 33
2-7 角點偵測方法 36
2-8 K-MEAN方法 44
第三章 提出的方法 50
3-1 研究方法介紹 50
3-2 研究方法流程步驟 53
第四章 實驗結果 60
4-1偵測實驗系統架構環境 60
4-2 偵測實驗影片資料庫樣本介紹 61
4-3 偵測實驗結果 67
第五章 結論及未來發展 74
5-1 本論文之貢獻 74
5-2 未來研究 74
參考文獻 76

[1] http://www.nfa.gov.tw/Upload/1/0307_105年全國火災統計分析_po上署網.pdf, 行政院內政部消防署pp.1-24, 2017
[2]http://www.nfa.gov.tw/Uploads/1/101-105年全國火災次數、起火原因及火災損失統計表xls, 行政院內政部消防署, 2017
[3]http://www.epochtimes.com/b5/17/10/10/n9718140.htm, 大紀元電子報, 2017-10-11
[4]https://www.theguardian.com/us-news/2017/oct/09/california-wildfires-wine-country-blaze, 2017 Guardian News and Media Limited,October 09, 2017
[5]http://calfire.ca.gov/communications/communications_StatewideFireSummary, California Statewide Fire Summary, October 28, 2017
[6]James G Quintiere, Principles of Fire Behavior. Delmar,1997.
[7]T. H. Chen, C. L. Kao and S . M. Chang, “ An Intelligent Real-Time Fire-Detection Method Based on Video Processing,” in Proceedings of the IEEE 37th Annual (2003) Inlernalional Carnahun Conference on Securiw Technology, pp.104-I 11,2003
[8]R.C. Gonzalez and R.E. Woods, Digital Image Processing 3rd ed., Pearson Education, Inc., 2008
[9]Ivan Kopilovic, Balazs Vagvolgyi, Tamas Sziranyi, “Application of Panoramic Annular Lens fur Motion Analysis Tasks: Surveillance and Smoke Detection,” IEEE Intemational Conference on Pattern Recognition., pp. 714-717.2000.
[10]Y. Cappellini, L. Mattii, and A. Mecocci, “An Intelligent System For Automatic Fire Detection in Forests,” IEEE 3th International Conference on Image Processing and its Applicalions., pp. 563-570, 1989.
[11]Wen-Bing Homg and Jian-Wen Peng, ‘‘ Real-Time Fire Detection From Video: A Preliminary Report,” 14th IPPR, Computer Yision, Cruphics ond Image Processing., pp. 1-5.2001.
[12]Yu, I., Wanh, N., &Meng, X. (2005). Real-time forest fire detection with wireless sersor networks. In Proceedings. 2005 international conference on wireless communications, networking and mobile computing, 2005.: Vol. 2 (pp. 1214-1217) IEEE.
[13]Ko, B. C., Cheong, K.-H., &Nam, J.-y. (2009). Fire detection based on vision sensor and support vector machines. Fire Safety Journal, 44(3), 322-329.
[14]Gollner, M. J. (2016). Detection and Suppression of Fires：A Cornerstone of Fire Protection Engineering. Fire Technology, 52(5), 1192-1196.
[15]Shen, P., Zhang, L., Song, j., Xu, H., Zhou, L., Wei, W., & Zhang, W. (2013). Dm642-based fire detection in video sequences using statistical color model. International Journal of Digital Content Technology and its Applications, 7(3), 699.
[16]Fleury, A. d. T., Trigo, F. C., & Martins, F. P. R. (2013). “A new approach based on computer vision and non-linear Kalman filtering to monitor the nebulization quality of oil flames”. Expert Systems with Applications, 40(12), 4760-4769.
[17]https://en.wikipedia.org/wiki/Additive_color, Wikipedia.org
[18]Z. Lin, J. Jin and H. Talbot, “Unseeded region growing for 3D image segmentation,” ACM International Conference Proceeding Series, vol. 9, pp. 31-37, 2000.
[19]https://en.wikipedia.org/wiki/YCbCr, Wikipedia.org
[20]https://en.wikipedia.org/wiki/Region_growing, Wikipedia.org
[21]Shilpa Kamdi, R.K.Krishna, “Image Segmentation and Region Growing Algorithm” International Journal of Computer Technology and Electronics Engineering (IJCTEE) Volume 2, Issue 1.
[22]R. Adams and L. Bischof. “Seeded region growing”. IEEE Trans. Pattern Anal. Mach. Intelligence, 16(6):641–647, 1994.
[23]https://en.wikipedia.org/wiki/Corner_detection, Wikipedia.org
[24]Harris and M. Stephens (1988). "A combined corner and edge detector". Proceedings of the 4th Alvey Vision Conference. pp. 147–151.
[25]H. Moravec (1980). "Obstacle Avoidance and Navigation in the Real World by a Seeing Robot Rover". Tech Report CMU-RI-TR-3 Carnegie-Mellon University, Robotics Institute.
[26]Mohanna F. , Mokhtarian F. 2001 “Performace Evaluation of Corner Detection Algorithms under Similarity and Affine Transforms.” Proceedings of the British Machine Vision Conference 2001
[27]Rosten E. and Drummond T. “Machine learning for high-speed corner detection”, ECCV 2006: Computer Vision – ECCV 2006 pp 430-443
[28]M. Piccardi (October 2004). Background subtraction techniques: a review (PDF). IEEE International Conference on Systems, Man and Cybernetics. 4. pp. 3099–3104.
[29]https://en.wikipedia.org/wiki/Background_subtraction, Wikipedia.org
[30]D. Stanin Alex, Dr. Amitabh Wahi, BSFD: Background Substraction Frame Difference Algorithm For Moving Object Detection And Extracdtion, JATIT, 28th Feb 2014. Vol. 60 No.3, pp623-628
[31]Nishu Singla, Motion Detection Based on Frame Difference Method, International Journal of Information & Computation Technology. ISSN 0974-2239 Volume 4, Number 15 (2014), pp. 1559-1565
[32]Kriegel, Hans-Peter; Schubert, Erich; Zimek, Arthur (2016). "The (black) art of runtime evaluation: Are we comparing algorithms or implementations?". Knowledge and Information Systems. 52: 341–378.
[33]https://en.wikipedia.org/wiki/K-means_clustering, Wikipedia.org
[34]J. B. MacQueen (1967): "Some Methods for classification and Analysis of Multivariate Observations", Proceedings of 5-th Berkeley Symposium on Mathematical Statistics and Probability, Berkeley, University of California Press, 1:281-297
[35]Bilkent EE Signal Processing Group, " Sample Fire and Smoke VideoClips,"[Online].Available: http://signal.ee.bilkent.edu.tr/VisiFire/Demo/SampleClips.html
[36]T. Celik and K.-K. Ma, "Computer Vision Based Fire Detection in Color Images," in IEEE conference on Soft Computing in Industrial Applications (SMCia/08), Muroran, Japan, 2008.