基于中心提取的RHT在椭圆检测中的应用

充分利用椭圆的几何性质,借助椭圆的形状控制点约束和弦端点法向约束,大幅降低随机Hough变换(RHT)的无效采样和累积次数,并采用基于视觉感知聚类的模糊置信度对由同一个形变椭圆引入的多个虚假候选椭圆进行有效去除.实验结果表明:该算法与基于RHT的其他椭圆检测方法相比,具有检测速度快、精度高、抵抗椭圆的部分缺失和形变能力强等优点.

Ellipse Detection by the RHT Based on Center Extraction

The useless samplings and accumulations of randomized Hough transform (RHT) are largely reduced with the help of the ellipse's shape control point constraint and the constraint of normals at the endpoints of a chord. The false candidate ellipses introduced by the same distorted ellipse were eliminated effectively using the fuzzy confidence based on vision perceptual grouping. The experimental results show that compared with the other ellipse detection method based on RHT, our proposed method has the advantage of higher detection speed and accuracy as well as strong resistance to ellipse's partial occlusion and deformation.