Large Sky Area Multi-object Fiber Spectroscopic Telescope – LAMOST, with a 1.75m-diameter focal plane on which
4000 optical fibers are arranged, is one of major scientific projects in China. During the surveying process of LAMOST,
the optical imaging system makes the astrometric objects be imaged in the focal plane, and the optical fiber positioning
system controls the 4000 fibers to be aligned with these objects and obtain their spectrum. In order to correct the
positioning error of these optical fibers, the CCD camera is used to detect these fibers’ position in the way of close-range
As we all know, the calibration quality of the CCD camera is one of the most important factors for detection precision.
However, the camera calibration has two following problems in the field work of LAMOST. First, the camera parameters
are not stable due to the changes of on-site work environment and the vibration during movement. So, the CCD camera
must be on-line calibrated. Second, a large-size high-precision calibration target is needed to calibrate the camera, for the
focal plane is very big. Making such a calibration target, it is very difficult and costly. Meanwhile, the large calibration
target is hard to be fixed on LAMOST because of the space constraint.
In this paper, an improved bundle adjustment self-calibration method is proposed to solve the two problems above. The
results of experiment indicate that this novel calibration method needs only a few control points while the traditional
calibration methods need much more control points to get the same accuracy. So the method could realize the on-line
high-precision calibration of CCD camera for LAMOST.