Improving space debris detection in GEO ring using image deconvolution

In this paper we present a method based on image deconvolution to improve the detection of space debris, mainly in the geostationary ring. Among the deconvolution methods we chose the iterative Richardson–Lucy (R–L), as the method that achieves better goals with a reasonable amount of computation. For this work, we used two sets of real 4096×4096 pixel test images obtained with the Telescope Fabra-ROA at Montsec (TFRM). Using the first set of data, we establish the optimal number of iterations in 7, and applying the R–L method with 7 iterations to the images, we show that the astrometric accuracy does not vary significantly while the limiting magnitude of the deconvolved images increases significantly compared to the original ones. The increase is in average about 1.0 magnitude, which means that objects up to 2.5 times fainter can be detected after deconvolution. The application of the method to the second set of test images, which includes several faint objects, shows that, after deconvolution, up to four previously undetected faint objects are detected in a single frame. Finally, we carried out a study of some economic aspects of applying the deconvolution method, showing that an important economic impact can be envisaged.