Research On Automatic Identification Of Space Motion Targets Under Starry Background

With the development of science and technology and the growing emphasis on space security,more and more space targets have been sent into space.Space is also important for the security of every sovereign nation.According to the state’s observation demand of space target,this paper studies the autonomous recognition technology of space target,which is helpful to improve the recognition ability of space target,and provide technical support for space target detection,space debris cleaning and on orbit maintenance service,and lay a solid technical base for the implementation of the project in the future.This paper studies the main characteristics of the space target,including the radiation characteristics,geometric characteristics,motion characteristics,and observability of the space target.And the spatial background characteristics are analyzed,then the difference of the motion characteristics between the background stars and the space target is theoretically calculated.According to the distance between the target and the observation satellite,a corresponding positioning method is proposed.When the space target is observed from a long distance,according to the difference of the motion characteristics between the target and the background stars,a point target detection algorithm based on Hough transform is proposed.If the target is a failure satellite,it will appear as a rotating target.When the target is observed for a long time,its imaging gray is changing regularly,so the rotation speed of the target can be inferred according to the change period of the target gray.The combination of vector angle and vector relative entropy is used to calculate the target rotation period.When the target is observed at a close distance,the target occupies most of the field of view,and the target can be positioned according to the docking ring on the target.The docking ring is generally concentric ellipse when imaging,so a Hough transform based on edge extraction is proposed.The ellipse detection method realizes detection of concentric ellipses to locate the position of the body.This paper simulates the above three algorithms on VS2015,Matlab and other platforms.The results show that all three algorithms are effective and feasible.In order to verify the effectiveness of the algorithm,this paper designs a set of high-speed digital image processing platform.The hardware platform mainly uses the structure of CMOS+FPGA+DSP to realize high-speed data processing.CMOS is mainly used for imaging the target.FPGA is mainly used to configure CMOS and DSP,preprocess image data and flow control,and control power-up sequence.DSP mainly implements image segmentation algorithm,centroid positioning algorithm,image registration algorithm and spatial target extraction algorithm.This paper uses the method of projection simulation to provide the data source of the detector,and uses the hardware platform to verify the spatial point target recognition algorithm.The experimental results show that the designed hardware platform is stable and the algorithm used is feasible and meets the engineering requirements.