Flame height is a key parameter which describes the flame characteristic and flame combustion condition. The most popular method to detect it is ocular estimate at present. As weak flame has characteristics of low bright,low signal-to-noise ratio and vague outline,ocular estimation won't work anymore,and the conventional image analysis method is highly depends on the threshold value. In order to solve these problems,after extracting incomplete flame edge by using the convention image processing method, multi-parameter model of flame contour is established to calculated flame height. The experimental result indicate that this algorithm has strong anti-jamming ability. Compared with the conventional method, the algorithm has character such as high precision, low dependence of threshold value and good consistency.%火焰高度是描述火焰特性、确定火焰燃烧状态的重要参数,目前多使用目测方法进行检测.微弱火焰具有亮度低、信噪比低、火焰轮廓模糊的特点,人工目测困难,常规图像分析方法检测对阈值依赖性高.为解决上述问题,在利用常规图像处理方法提取的不完整火焰边缘后,建立火焰轮廓的多参数模型,进而计算火焰高度.实验结果表明:该算法抗干扰能力强,相比常规方法具有精度高、对阈值依赖性低、一致性好的特点.
展开▼
