A Survey on Image Processing based Techniques for Space Debris Detection

Abstract

Space debris is the nonfunctional objects revolving in the outer earth orbit. The high-speed debris projecting to Kessler's syndrome is a threat to active satellites. Rigorous attempts are being made for the efficient detection of the debris for safeguarding existing functional satellites. Among them, image processing methods prove beneficial for the detection of space debris in orbit. The image captured from the satellite is in the form of visible images and thermal images. The survey is conducted on various image processing techniques for space debris detection based on ground-based tracking, satellite-based, simulation-based, and fusion-based detection. The features of debris detected by the various methods are thoroughly reviewed. The study enhances the importance of the fusion of the features of visible and thermal images provided with Deep Neural Networks (DNN). The fusion-based method provides an efficient solution to detect debris in both sunlit and non-sunlit areas from the satellite.