视频火灾检测的设计与实现

Design and implementation of fire detection with video

在分析火灾图像特性的基础上,运用数字图像处理技术和模式识别技术,提出了火灾识别的思想。给出了图像处理和识别的算法,该算法采用二维最大熵自动阈值法对火灾图像进行分割处理,分割后再提取可疑区域;对可疑区域的火焰进行识别,给出火焰存在的可能性;根据火灾火焰蔓延时的面积、相似度的变化来识别、判断火灾的发生。实验证明,与传统的检测方法相比,大大地提高火灾预报的准确率。

The idea of fire recognition is introduced based on analyzing fire image characteristic using digital image processing technology and pattern recognition technology. The image processing and the recognition algorithm about the fire are presented. This algorithm is as follows. Firstly, two-dimensional maximum-entropy automatic threshold is used for segmentation of the fire image, after which the suspected light-area in the image is distilled. And then, the recognition of flame is done in the suspected light-area, from which the possibility of the flame existence is gained. Lastly, it can judge whether fire occurs from the fire flame spreading area and the similarity change. The experimental results show that the method can greatly improve veracity of fire prediction, comparing with the traditional detection method.