基于中点提取的椭圆检测算法

针对Hough变换检测椭圆时存在运算速度慢、需要大量储存空间的问题,提出一种快速的随机椭圆检测算法。该算法随机采样2点并搜索第3点,再利用椭圆极径性质求出椭圆中心坐标,接着通过坐标变换得到简化的椭圆方程,解方程组求出椭圆剩下的参数。在证据收集确认候选椭圆时,只选取椭圆外切矩形内的边缘点,同时利用特殊的累积记录函数对候选椭圆参数进行进一步判别。实验结果表明,该算法具有检测速度快、精度高、抗椭圆部分缺失的能力强等优点。

Ellipse detection algorithm based on center extraction

For the Hough transform method for ellipses detection is slow and needs huge storage, a fast randomized algorithm was presented. The algorithm randomly sampled two edge-points and searched for the third point to determine the center coordinate of an ellipse by the characteristics of pole chord, and then solved the elliptic systems obtained by coordinate transformation to get the rest parameters of the ellipse. At the time of determining a candidate ellipse by a verification process, only the points in the enclosing rectangle of the ellipse were selected. And the authenticity of a potential ellipse was confirmed by a special cumulative function. The experimental results demonstrate that the proposed method has the advantages of high speed and accuracy as well as strong resistance to partial occlusion.