摘要
视觉测量系统中圆形标志点中心定位的精度易受噪声的影响。为了增强其抗噪性从而提高定位精度,提出了一种利用几何特征以降低噪声干扰的中心定位算法。首先将自适应阈值分割法与质心法相结合,对点目标进行圆心粗定位。利用粗定位的圆心和半径对Canny算子检测到的边缘进行半径约束,以消除孤立点和噪声点。然后根据理想的圆成像后边缘点分布的几何特征和链接规律,采用一种基于分区原理的方法获取点目标的理想边缘。最后,采用Zernike正交矩对像素级边缘点进行亚像素定位,并用最小二乘椭圆拟合法计算得到中心坐标。实验结果表明,该方法的定位精度可以达到0.023 7pixel,算法的运行时间为2~3ms,基本满足测量系统对于圆形标志点中心定位在精度、稳定性和实时性上的要求。
The location accuracy of circle target center in vision measurement system is susceptible to the noise.In order to enhance the capacity of anti-interference of noise and improve the location accuracy,a modified subpixel location algorithm of circle target center based on geometric features is proposed in this paper.Firstly,coarse location of circle target center is realized by combining adaptive threshold segmentation and centroid method,which is used for radius constraint of the edge detected by Canny operator to remove the isolated points and the noisy points.Then,starting from the geometric distribution characteristics and linking discipline of the ideal imaging of circle,a method based on partition theory is brought forward to obtain ideal and continuous edge of circle target.Finally,Zernike moment is used to carry out the subpixel location of the edge,and the center location is achieved by the least-squares ellipse fitting method.The experiments results show that the locationaccuracy can reach to 0.023 7pixels,and the operation time is between 2and 3ms.It can satisfy the high requirements of precision,stabilization and real-time ability of circle target center in the measurement system.
出处
《液晶与显示》
CAS
CSCD
北大核心
2014年第6期1003-1009,共7页
Chinese Journal of Liquid Crystals and Displays
基金
973项目(No.2013CB733103)
