A New Metrology Method for Optical Fiber Position Detection

Mengtao Li,Jiale Zuo,Shipeng Duan,Zhigang Liu,Hongzhuan Hu,Jianping Wang,Jiaru Chu,Zengxiang Zhou
DOI: https://doi.org/10.1117/12.2561056
2020-01-01
Abstract:The entire system of the LAMOST ( Large Sky Area Multi-Object Fiber Spectroscopic Telescope) requires high positioning accuracy of the fiber positioning unit. In order to acquire accurately target celestial objects, fiber view metrology system for positioners can efficiently and accurately detect thousands of fiber spots simultaneously in a large scale is required. The traditional method mainly used the "back-illumination method" for detection. With the advent of 8k*6k high-resolution CMOS cameras, fiber position detection based on the "front-illumination method" becomes feasible. This paper mainly studies the fiber position detection based on the "front-end illumination image processing method". The image is preprocessed first, and then the edge detection of a large number of fiber target points in the image is performed. Considering the constant radius of the white ceramic head where the fiber is located, the article proposes a "front-illuminated" image algorithm based on radius-based Hough space conversion and optimal radius error center search. This algorithm improves the speed and accuracy of fiber pixel coordinate detection. At the same time, it can be coordinated and compared with the "back-illuminated method" to further optimize and improve the detection accuracy of the fiber position.
What problem does this paper attempt to address?