Space Debris Automation Detection and Extraction Based on a Wide-field Surveillance System

Wide-field telescopes with long exposure times have stronger space target detection capabilities. However, complex background sky conditions will still cause a series of difficulties in detecting space debris, such as a large number of star points, a large amount of noise, and the discontinuity and nonlinearity of the target. We propose a space debris automatic extraction channel with a high detection rate and low computational cost to solve these difficulties. We apply an improved median filter for noise elimination and then the double-structure morphological filter algorithm used to suppress the background of the star image to eliminate star points and noise. Then, the guided filter was used to eliminate residual noise, and star points were used to reduce the impact on the target. Finally, the improved Hough transform was also applied to detect the target in the image. Our automatic extraction algorithm is used in real astronomical star maps, including different orbiting satellites (star-tracking mode). These images were obtained by using a 280 mm diameter telescope, which was located in Changchun Observatory. The experimental results demonstrated the effectiveness of the extraction algorithm in this study. It can effectively detect and track space targets in a long-exposure wide-field surveillance system and has high positioning accuracy and low computational complexity, which solves the problem of space debris extraction under a complex background.