摘要
提出一种燃烧器射流流场分析及几何特征参数提取的方法,实现流程包括:①采用加权中值滤波算法抑制射流图像离散点噪声;②采用种子区域生长算法将射流从图像背景中分割;③采用自适应边缘检测算法检测射流边缘像素点;④采用广义最小二乘法对射流边缘点进行线性拟合;⑤确定射流的几何中心线位置、中心角及射流扩展角等特征参数.研究结果表明,当处理图像噪声窗口为10×10,选择射流图像中亮度值最高的像素为种子,同时以5%为误差限确定生长准则时,整个流程消耗时间与精度处理的匹配最优.该燃烧器流场的实时测量与分析的方法具有较好的应用前景.
An algorithm for flow-field analysis and geometric feature extraction of single burner＇s jet is presented.It includes the following steps： ① discrete-points denoising based on weighted mean filter algorithm;② image segmentation based on region growing algorithm;③jet edge pixels acquiring based on adaptive edge detection algorithm;④ linear fitting based on generalized least square method;⑤ locating the centerline position of the jet and extracting other geometric parameters.Research results show that time consumption and calculation precision will be well satisfied for a given filter operator of 10×10 and an image tolerance of 5%.The developed algorithm of real-time measuring and analysis for the air flow field of burner jet has a good application prospects.
出处
《电力科学与技术学报》
CAS
2007年第3期41-45,共5页
Journal of Electric Power Science And Technology
基金
湖南省教育厅科研资助项目(06C130)
关键词
图像处理
图像分割
边缘检测
特征提取
燃烧器射流
Image processing
Image segmentation
Edge detection
Feature extraction
Burner＇s jet
