基于随机抽样与投票策略的缺陷圆检测方法

Defect circle detection method based on random sampling and voting strategy

为了实现缺陷圆圆心与半径参数的检测,本文提出一种基于随机抽样与投票策略的检测方法。通过在目标轮廓上随机提取三点并由此获得任意两条弦,计算它们各自对应的垂直平分线的交点。重复上述过程可得到若干候选圆心坐标,应用投票策略检测出频数最高的坐标值,得到理想圆心。在此基础上再根据投票策略获取半径参数。仿真和实验结果表明,本文方法适应性强,鲁棒性好,可应用于不同类型缺陷圆检测;其检测结果准确可靠,易于实现,可以满足工程应用需求。

To detect the center and radius parameters of defect circle,the paper proposes a detection method based on random sampling and voting strategy.Through randomly selecting three points from the target edge,two arbitrary chords are obtained,and the intersection points of two vertical bisectors are calculated.Several candidate center coordinates are obtained by repeating the above process.By using voting strategy,one can extract the coordinate value holding the highest frequency and so the ideal circle’s center is detected.Then the radius parameter is obtained by using the voting strategy again.The results of simulation and experiment show that the proposed method has strong adaptability and robustness and may be applied to different types of defect circle.In addition,it is accurate and reliable,easy to realize,and can meet the requirement of engineering application.