Research Of Infrared Dim And Small Target Detection Technology On Space-based System

Since the Second World War,ballistic missile technology has developed rapidly,which leads to the emergence of anti-satellite missile technology.The Exo-atmospheric Kill Vehicle(EKV)is one of the most representative ones.With the rapid development of anti-satellite missile technology,each military power began to establish ballistic missile defense system,so the space-based infrared dim and small target detection system came into being.Compared with radar detection system,the passive detection and anti-interference advantages of infrared detection system make it be irreplaceable on the space-based platform.Because the low-orbit satellites of space-based infrared system are easily be attacked by EKV,the detection and tracking of EKV is of great significance.Based on the project of ‘Long-range High-speed Infrared Target Search Syste',infrared dim and small target detection algorithm under starry background is proposed and performed on hardware system based on DSP&FPGA to achieve real-time and accurate detection for dim and small targets including EKV.In order to achieve high-performance detection of EKV,infrared radiation characteristics of EKV is studied at first,and the thermal equilibrium equation is established to obtain its temperature characteristics.Secondly,EKV's motion model is established through deep analysis to its motion characteristics.Then,environmental features and noise characteristics of starry background are studied.As a result,the main objects that can be captured by the imaging system include stars,satellites,space debris,targets and various noises.Minimum star radiance and the number of stars detected by the infrared system are calculated according to the data information of stars in the catalog.The motion models of satellite and space debris interference are established and the Gaussian model is adopted to simulate various noises.Finally,the software for simulating moving and imaging of space-based platform is established by combining all the models mentioned above.By setting the initial position of satellite and space debris,images containing infrared dim and small target under starry background are generated.Infrared dim and small target detection algorithm under starry background plays a very important role in target detection and tracking system.Based on the analysis of advantages and disadvantages of traditional single-frame detection and multi-frame detection algorithm,an infrared small target detection algorithm with sky background is designed which combines detection algorithms based on single frame image with one based on image sequences.This algorithm includes background suppression,threshold segmentation,candidate target marker,feature points extraction,map matching and target confirmed.In this algorithm,the regional mean background suppression method is used for background suppression,the global threshold computed by mean and variance of images is used for threshold segmentation,the weighted centroid method is used for marking.In order to facilitate multi-star point matching and star point removal,the triangle algorithm can be adopted.Simulation results show that the proposed method can remove stars and detect dim and small targets effectively.Infrared dim and small target detection algorithm under starry background is implemented on the hardware platform based on FPGA+DSP.The model of FPGA and DSP are respectively XC5VLX330 T,and TMS320C6678 with multi-core.The test results of the hardware system demonstrate that dim and small target detection algorithm in this paper can be implemented real-timely with high detection performance.