莫尔圆的特征参数提取方法

A Method for Recognizing of Moiré Circle Characteristic Parameters

在超精密测量领域中 ,通过处理同心圆光栅副迭合的莫尔条纹图像可以分离出二自由度的微小误差信息。此前需要对莫尔条纹图像进行预处理 ,提取出莫尔圆的特征参数。本文讨论了莫尔条纹图像处理的算法问题 ,介绍了图像预处理算法 ,包括图像增强、图像滤波、边缘检测、二值处理、轮廓细化、剔除野点等过程 ,最终用最小二乘法实现对莫尔圆的圆心位置与半径三个特征参数的提取。所采用的一系列方法使单幅含CCD噪声图像的莫尔圆特征参数的定位精度达到亚象素级。

In the field of ultra-precision measurement, tiny two-degree-of-freedom error information can be separated from moiré fringe image generated by overlapping a pair of concentric-circle gratings. Before this process, moiré fringe image need to be pre-processed to get moiré circle characteristic parameters. This paper discusses the algorithm problems on moiré fringe image processing and introduces pre-processing algorithms including processes of image enhancement, image filtering, edge detection, binary management, profile thinning, wild points erasing etc. Least square method is finally used to get the center and radius of moiré circles. The series of methods for recognizing of single image with CCD noise mixed have an accuracy of sub-pixel level.