Research On The Processing Method Of CCD Star Map In Wide-Field

As a kind of astronomical geodesy instrument,photographic zenith telescope usually has a small field of view,the field of most instruments is less than 1°.At present,the field of the photographic zenith telescope that has been put into observation is about 1°,Because the large field can take more stars,and then improve the accuracy of the solution of the plumb line of the observation site,the field of the new photographic zenith telescope increased to 2°.34.as the field of the new photographic zenith telescope increases,the aberration,field distortion and projection error will further increase.The ideal tangent plane projection method for small field needs to be improved.This paper analyzes and compares many kinds of star map processing methods,uses the actual star map of the photographic zenith telescope to explores the processing method of CCD star map in wide-fieldStar map processing methods are mainly divided into star image centroid extraction and star map identification.Star image centroid extraction is to obtain highprecision star image centroid position.Star map identification is a process of feature matching based on the position information of star image and reference star.In the star image centroid extraction,after analyzing and comparing the effect of Gauss filter and median filter,it is considered that Gauss filter is more suitable for CCD star image processing;according to the gray characteristics of star image and noise,star image region is extracted by threshold segmentation;in the process of star image centroid extraction,different centroid calculation methods are compared,and the results show that the weighted centroid method is more stable and efficient.In the work of star map identification,triangle algorithm is selected for star map identification.The centroid position of the star image is measured in the twodimensional film coordinate system,and the position of the reference star is described in the three-dimensional spherical coordinate system.Before using the triangle algorithm to recognize,we need to transform the coordinate system,one is to convert the reference star to the ideal tangent plane and then use the triangle algorithm for identification,for the ideal tangent plane projection,the projection formula method and MDS algorithm are used respectively.The other is to convert the star image to the threedimensional photographic zenith telescope coordinate system and then use the triangle algorithm for identification.Finally,by using the observation data,the comparison of various identification schemes shows that the scheme which first transforms the star image to the threedimensional photographic zenith telescope coordinate system and then uses the triangle algorithm for identification can obtain better position resolution accuracy.