【摘要】 提出了一种在视觉检测中用于多圆提取的改进随机Hough变换算法,该算法采用一种新的取点方法来进行计算,避免了随机Hough变换由于随机采样造成的大量无效累积。并在计算圆的几何参数时充分利用了圆的几何特征,因此检测速度快,占用内存空间小,精度高,具有良好的抗噪和抗干扰能力。最后文中还分别给出了人工合成图和实物图的实验结果。更多还原

【Abstract】 An improved randomized Hough transform (RHT) algorithm is proposed for detecting multi-circles in vision inspection. This algorithm uses a new way to get points for calculating the parameters of the circle which avoids the mass useless accumulation caused by RHT,and it uses the geometric features of the circle a lot when calculating the parameters. The algorithm is fast,high precision and takes only a little storage space,performs well against noise as well. At the end of the paper,it also gives the experi-mental results of synthetic image and real image.更多还原