Methods Accurately Locating Optical Positions of Space Object Based on CCD Stellar Images

Optical measurements of positions of space objects are important means to derive the orbits of moving space objects for surveillance.We introduce a miniature system of locating space objects optically,which is based on recording CCD images.We apply denoising methods of Top-hat transformation and median filter to a CCD image recorded for an object.We further use a new algorithm based on the Grads Hough Transformation for simultaneously and accurately locating the positions of background stars the track of the object.Finally,we are able to accuretly derive the position of the space object based on invariance of angular separations between stars and formulas of spherical geometry.The results of applying the system and methods to a satellite with size of the CCD image 752×580 show that the methods of image denoising and the algorithm of locating stellar positions and object track are effective,and the accuracy of locating the position of a moving space object can be of a few arcseconds.