一种快速检测圆心的抗噪声亚像素算法

A Robust Subpixel Algorithm for the Center of Circle Detecting

在印刷电路板的自动光学检测中，被检对象的空间对准是一个关键步骤。而传统的图像匹配技术由于其只能在像素级定位，而无法适应印刷电路板精确对准的要求。本文利用圆的几何对称性，提出一种在亚像素精度快速定位圆心的算法。实验表明：它对于无噪声图像的测量误差小于０．０１像素；在有噪声的情况下，该算法仍具有较小的测量误差。

The part alignment is an important step for automated optical inspection of printed circuit board (PCB). Because conventional image recognition algorithm could only be implemented at pixel accuracy, it is not satisfied with the requirements of PCB inspection. Through the use of geometric symmetry of a circle, a robust subpixel algorithm for the center of circle detecting is presented. By the proposed method the measurement error might be less than 0.01 pixel for noise free images. In the meantime, experiments show that the algorithm has a relatively small bias in the presence of noise.