转载请注明出处:http://blog.csdn.net/xiaowei_cqu/article/details/7523683
有关火灾检测的几篇论文的算法小总结:
An Early Fire-Detection Method Based on Image Processing
Thou-Ho (Chao-Ho) Chen, Ping-Hsueh Wu, and Yung-Chuen Chiou In: Proc. IEEE Internat. Conf. on Image Processing, ICIP 04, pp. 1707–1710.
在
森林火灾检测-1实现的检测初级版本中,用的就是这篇论文的颜色模型,论文整体结构也很简单:
![](https://img-my.csdn.net/uploads/201204/30/1335756143_4525.png)
Computer vision based method for real-time fire and flame detection
Töreyin, B., Dedeoglu, Y., Gudukbay, U., Cetin, A.,Pattern Recognition Letter, Vol. 27, pp. 49-58, 2006.
这也是引用很多的一篇。前面两部也是通过运动模型和颜色模型,之后综合使用一个高斯分布模型:
![](https://img-my.csdn.net/uploads/201204/30/1335756327_9968.png)
Fire detection based on vision sensor and support vector machines
B.C.Ko,K.H.Cheong,J.Y.Nam,FireSaf.J.44(2009)322–329.
这是韩国Keimyung大学CVPR实验室(Computer Vision and Pattern Recognition Laboratory, Keimyung University, Korea)做的,相应做了四个版本。在
http://cvpr.kmu.ac.kr/body.html#FireDetection可以看到他们各个版本的介绍以及演示效果。对比四个版本,低级profile及用运动和颜色的模型都是相似的,高级profile有较大变化:
![](https://img-my.csdn.net/uploads/201204/30/1335757013_3000.png)