Research on the Fiducial Fibre Coding and Identification Algorithm of the LAMOST Fibre Positioning System

Abstract

To obtain the world coordinates of a measured object in camera measurement, a certain number of image control points should be established around the detected object as fiducial points for data processing. The precision of a fiducial point directly determines the measurement accuracy of the measured object in the camera measurement. Fiducial points in camera measurement are widely used in various fields of measurement, such as aerial surveys, close-range surveys, and 3D reconstruction. In addition, a fiducial point generally adopts an encoding structure to reduce the search cost in the detection. Camera measurement is used in the LAMOST fiber positioning system, which also requires fiducial points to ensure the accuracy of fiber positioning. Different from a general fiducial point, the whole system needs to operate under dark conditions during LAMOST observation, and optical fiber positioning is carried out by the “backlighting method”, which measures the optical fiber spot under a dark background. In addition, fiber positioning needs to reach an accuracy of 40 μm at a distance of 20 m, which is difficult to achieve with the fiducial points used in general vision measurements. To meet the requirements of LAMOST optical fiber positioning with high accuracy, this paper presents a fiducial fiber and its recognition algorithm for optical fiber positioning of optical fiber spectroscopic telescopes. The fiducial fiber can be applied to a wide-field optical fiber spectroscopic telescope measurement system in a high-precision long-distance measurement environment. The experimental results show that fiducial fiber can ensure the accuracy of closed-loop positioning for LAMOST fiber units.