基于视频的火焰检测技术研究  

Flame-Detection Technique Based on Video Processing

作　　者：高爱莲[1] 刘辉[1] 林宏[2] 王汉友[3] 

机构地区：[1]昆明理工大学信息工程与自动化学院,昆明650051 [2]西南林学院计算机与信息科学系,昆明650224 [3]昆明理工大学机电学院,昆明650093

出　　处：《云南民族大学学报（自然科学版）》2008年第1期94-96,共3页Journal of Yunnan Minzu University:Natural Sciences Edition

摘　　要：综合分析了火焰的颜色、闪烁频率及其面积增长性等静态特征和动态特征,设计了一种基于视频的火焰检测自动预警算法.利用混合高斯模型有效地表示了火焰颜色及颜色的多态性.实验表明该算法具有误报警低、识别率高、实时性强等优点,是火灾自动报警的一种可行、科学的解决方案,具有广泛的应用前景,该研究在国内外还处于发展阶段.The paper analyzes static characteristics and dynamic ones of flame such as flame＇s colors,flickering frequency and augmentation, puts forward a new flame - detection algorithm that could automatically give an alarm on video processing. Flame＇ s color and its polymorphism were described effectively by mixture Gauss model. The experimental results show that the flame - detection algorithm has many advantages such as lower false alarm rate, higher recognition rate and it can be used in many real - time control environments. It is an efficient and scientific solution, and it has widely applied prospect. The research of flame - detection is stiU in the early stage in China.

关 键 词：火焰检测 混合高斯模型 静态特性 动态特性 

分 类 号：TP391.4[自动化与计算机技术—计算机应用技术]