论文部分内容阅读
针对传统圆检测算法检测速度慢、不适于多圆检测的问题,提出一种基于全局搜索的圆检测方法。将证据积累和加权平均的思想结合,对证据积累过程中产生的伪圆心进行归类、分析,并对三类伪圆心进行逐类剔除,最后计算其它圆参数。实验结果表明,该算法效率高,对局部信息缺损不敏感,检测时间不会随着圆个数的增加而线性增加,检测效果明显优于传统的随机圆检测(RCD)算法。
Aiming at the problem of slow detection speed and unsuitable for multi-circle detection in traditional circle detection algorithm, a round detection method based on global search is proposed. Combining the idea of evidence accumulation and weighted average, the pseudo-centroids generated in the process of evidence accumulation are classified and analyzed, and the three types of pseudo-centroids are eliminated class by class, and finally other circular parameters are calculated. The experimental results show that the proposed algorithm is efficient and insensitive to local information loss. The detection time does not increase linearly with the increase of the number of circles. The detection result is obviously better than the traditional random circular detection (RCD) algorithm.