基于凹凸变换与圆周拟合的重叠气泡轮廓重构

Contour reconstruction of overlapped bubbles based on concave-convex transformation and circle fitting

针对识别气液两相流的气泡过程中气相比例较大气泡易发生重叠的问题,提出基于曲线凹凸性变换与圆周拟合的重叠气泡分割与轮廓重构算法.对气泡图像进行预处理,利用高斯滤波滤除噪声干扰,依据自适应阈值分离出前景,得到预分割图像.利用Canny算法提取轮廓,利用高斯函数平滑曲线,计算轮廓上的点到质心的距离.对得到的距离曲线进行凹凸性变换特性分析,对每两个凹点进行分割路径选取.利用最小二乘圆周拟合插值算法对重叠部分的气泡轮廓进行重构,得到气泡分割后的重构图像.实验结果表明,该算法是一种有效的分割与重构重叠气泡的算法,具有较好的分割和重构性能.

A method for segmentation of overlapped bubbles segmentation and contour reconstruction was proposed based on the curve concave-convex transformation and circle fitting in order to identify overlapped bubbles of gas-liquid two-phase flow when the gas phase ratio is large.The bubble image was preprocessed,including the noise that was filtered by the Gauss filter,and the foreground that was separated according to the adaptive threshold.The pre-segmented image was obtained.Then the contour was extracted by the Canny algorithm.The curve was smoothed by the Gaussian function.The distance between contour points and the centroid was calculated.The curve was analyzed by concave-convex transformation characteristic.The segmentation path was selected on each of the two concave points.The bubble contour was reconstructed by the least square circle fitting interpolation algorithm,and the bubble segmentation and reconstruction image was obtained.Experimental results show that the proposed algorithm is an effective algorithm for the segmentation and reconstruction of overlapped bubbles,which has high performance.