基于改进最小二乘的缺陷椭圆定位方法

A Defect Ellipse Locating Method Based on Least Squares

分析形心法、Hough变换法和最小二乘拟合法等典型椭圆中心提取算法在应用中对噪声过于敏感、抗干扰性差的不足,提出了一种基于随机抽样一致性方法的改进最小二乘椭圆中心提取算法,经过仿真实验分析表明,该算法与最小二乘法、形心法相比具有更高的定位精度和强鲁棒性。

In this paper,an improved elliptic ellipse center extraction algorithm based on stochastic sampling consistencymethod is proposed,which is too sensitive to noise and has poor anti-jamming performance in typical ellipse center extraction algo.rithms such as centroid method,Hough transform method and least squares fitting method. Simulation results show that the proposedalgorithm has higher accuracy and strong robustness compared with the least square method and the centroid method.