基于分形方法的载人压力容器内壁缺陷可视化检测方法

Visual Inspection Method for Inner Wall Defects of Manned Pressure Vessel Based on Fractal Method

为提高高压氧舱内壁缺陷检测效率,降低漏检率,提出一种基于分形方法的高压氧舱内壁缺陷可视化检测方法。利用分形方法将数据点置于双对数坐标上,构建各阶累积和,选取最佳的分形参数作为缺陷参数。通过在高压氧舱内移动相机获取内壁上缺陷图像,计算图像中点的一阶差分,通过Canny算子求得各点梯度的幅值和方向,实现对缺陷轮廓的边缘检测;测量轮廓线长度和外接椭圆圆度拟合处理,并通过腐蚀运算和膨胀实现对缺陷的检测。仿真分析表明,该方法可有效检测出容器内壁缺陷,成像稳定,且对缺陷的检测偏差率较低,满足实际应用条件。

In order to improve the detection efficiency and reduce the high rate of missing detection,a visual detection method based on fractal method is proposed.By fractal method,the data points are placed on the double logarithm coordinates,the cumulative sum of each order is constructed,and the best fractal parameter is selected as the defect parameter.By moving the camera in the hyperbaric oxygen chamber,the defect image on the inner wall is obtained,the first-order difference of the midpoint of the image is calculated,the gradient amplitude and direction of each point are obtained by Canny operator,and the edge detection of the defect contour is realized.The length of the contour line and the roundness of the external ellipse are measured,and the defect is detected by corrosion operation and expansion.Through simulation analysis,the experimental results show that the proposed method can effectively detect the inner wall defects of the container,the imaging is stable,and the detection deviation rate of the defects is low,which meets the practical application conditions.