Research On Detection And Tracking Technology Of Space Target Observation Telescope

With the rapid development of space technology and the accelerated pace of human exploration of space,more and more space targets were sent into space,the outer space of the earth is gradually becoming a new military struggle.Under the new situation,whether the space can be effectively controlled determines the country's military power and national security.Therefore,it is of great significance to study the related technologies of the space target surveillance system.The main task of the space target monitoring system is to detect,track and identify the space targets,and to determine the type,size and trajectory of the target.It is the basis of entering,understanding and control space,and it will play a key role in the future space war.Aiming at the detection and tracking of space targets in complex starry sky background,this paper focuses on the design of the space target detection and tracking system,orientating the pointing of the telescope,the detection and tracking of moving objects in the background of the stars.The main contents of this paper are as follows:Firstly,aiming at the problem that the existing astronomical observation equipment does not meet the requirements of automatic observation of space target,this paper designs an automatic space target detection and tracking system.On the basis of the existing astronomical observation equipment,the system adds computer control,CCD image acquisition and the relevant adapter and so on.The system using a the large field of view camera to expand the observation range,so as to orient the pointing of the telescope and detect small space target.The system using small field camera observation object to obtain contour,gray and size feature information of space target,in order to keep track of the target.In addition,the hardware connection and software design of the system are given,and the technical difficulties are analyzed in this paper.Secondly,aiming at the problems of human error and complicated process in the manual orientating the pointing of the telescope at present,this paper proposed a new algorithm for automatic orientating the pointing of telescope.In this algorithm,we use the star map matching algorithm to extract the navigation stars in the real map and determine the celestial coordinates of the navigation star;then converted the celestial coordinates into the coordinates in the horizontal coordinate system;finally,using the camera attitude calculation model to orientating the instantaneous pointing of the telescope.In this paper,we used simulation experiment to test the proposed algorithm,verified the accuracy and efficiency of this algorithm.Then,aiming at the problems of the dim characteristics of the small targets,the space background interference and the noise of CCD image,this paper proposed a new algorithm of small target detection based on VIBE background modeling algorithm.The algorithm firstly use point target enhancement and background suppression algorithm to enhance the dim target on the star map,then use VIBE background modeling algorithm to detect small moving targets,finally use the method of object extraction and trajectory recognition to extract the trajectory of the space target.Experiments show that,compared with the original VIBE algorithm,the proposed algorithm can effectively detect dim moving targets on the star map,and can track the moving object extraction.Finally,aiming at the problem of losing the target in the field of large and small field of view switching and the rotation of the space object,the change of the edge contour and the gray level are difficult to be tracked in the small field of view,this paper proposed a target trajectory prediction algorithm based on Kalman filter and multiscale target tracking algorithm based on KCF.In the process of large and small field of view switching,the direction of the telescope is adjusted and the target is always in motion.In order to avoid the loss of the target,this paper uses the trajectory prediction algorithm based on Kalman filter to predict the position of the target.The experimental results show that the proposed algorithm can meet the system requirements for the prediction of the linear motion trajectory of a target in a finite frame.In the small field of view mode,this paper uses the multi-scale target tracking algorithm based on KCF to track the target with a certain contour.The experimental results show that the tracking algorithm can overcome the target rotation,scale and gray level changes,and achieve the goal of continuous tracking of the moving target.