考虑中心对称约束的标记点高精度定位方法

High-precision positioning of targets using visual methods has a wide range of applications in scientific research and industrial production. This paper studies the positioning algorithm of centrosymmetric marker points. Firstly, the marker points are arranged in a centrosymmetric array, providing robust positioning information for subsequent marker point positioning. Secondly, the center of the circular marker point is positioned using the method of Gaussian template fitting, which can effectively suppress the influence of noise, etc. Finally, the positioning result is optimized using the central symmetry of grayscale to determine the final position of the marker point. Verified by experiments, compared with the positioning method based on contour information, the applicable range of image contrast of the method in this paper is significantly improved. Compared with the traditional positioning method based on template matching, the method in this paper has better noise resistance and stability.

利用视觉方法对目标进行高精度定位在科研和工业生产中有着广泛的应用.本文研究了中心对称标记点的定位算法.首先,标记点的设置采用中心对称阵列排布的方式,为后续标记点定位提供鲁棒的定位信息;其次,利用高斯模板拟合的方法对圆形标记点中心进行定位,能够有效地抑制噪声等影响;最后,利用灰度的中心对称性对定位结果进行优化,确定最终的标记点位置.经实验验证,与基于轮廓信息的定位方法相比,本文方法对图像对比度的适用范围有明显的提升.与传统的基于模板匹配的定位方法相比,本文方法有更好的抗噪性和稳定性.