Early Forest Fire Target Detection Based On Statistical Pattern Recognition

With the development of social economy,there have been more and more high buildings and forest fire problem.The fire problem have such features as fire spreading fast, difficulty of extinguishing a fire and rescuing human beings and properties. So new-style and efective measure for the detection of early fire have been paid more attention for people. In this paper, by the aid of digital image processing and statistical pattern recognition techniques, an early fire image detection algorithms based on statistical pattern recognition have been designed and realized.The principle of fire occurring, developing stages and characters of every stage have been discussed in this paper. The difference is also studied between conventional fire detection systems and new-style fire detection techniques. On the base of the analysis of optical characters of fire smoke, the algorithms of fire image enhancement, noise filtering, image threshold segmentation, statistical pattern recognition techniques and their applications in the fire image proessing have been studied in this paper. In addition,by moving object detection, accumulate dynamic data,image segmentation and so on these process, the fire smoke image segmentation have been realized.At last,the method of fire recognition in the image by object similarity ,area and other characteristic have been discussed.The fire detecting simulation and the program procedure has been given in the last part of this paper. The experiment results show that this early fire image detection algorithms based on statistical pattern recognition can efficiently and reliably detect smoke with indoor background and not very complex outdoor background.