Subpixel edge detection of inner hole for ceramic optical fiber ferrules

摘要

Concentricity is the most important parameter to show the quality of the ceramic optical fiber ferrules. The diameter of its inner hole is micron. Therefore, the method of image detection is suitable to detect the edge of the inner hole. In this paper, we firstly introduce some common subpixel detection algorithms. Then three different kinds of edge gray models are presented. Finally, according to the gray scale characteristics of edges of inner holes, we put forward a subpixel edge detection algorithm by using the gradient fitting method based on the Gauss function. Besides, polynomial interpolation method and gray moment method are also used to detect the edge of the image. The experimental results show that: the gradient fitting method based on the Gauss function has the higher precision than other methods.