Date: September 2016
Authors: Hart, M., Jefferies, S., Hope, D., Nagy, J.
Abstract: High resolution daytime imaging of resident space objects (RSO) from the ground is presently severely challenging. At visible wavelengths, where diffraction-limited resolution is the highest before the atmosphere becomes opaque in the UV, shot noise from the bright background degrades the information that may be recovered from RSO imagery. Total exposure times must be limited in order to avoid motion blur induced either by the object’s intrinsic rotation or simply by its orbital motion over the site. Fundamentally, then, one cannot collect enough light from the object to achieve adequate signal-to-noise ratio (SNR) in the presence of very high noise before the apparent shape of the object has changed. To overcome this limitation, we propose in this paper a suite of techniques which we believe will collectively enable high-resolution imaging during daylight. The approach, which has yet to be fully implemented, relies on a sequence of short-exposure images from a high-cadence camera together with simultaneous wave-front sensor (WFS) measurements acquired from a filtered sodium laser guide star. We then directly estimate the three-dimensional shape of the RSO using a formalism similar to the concept of deconvolution from wave-front sensing (DWFS). In this way, provided that the intrinsic shape of the RSO does not significantly change during the course of the observations, we can combine data from quite different pose angles in order to achieve a high resolution result with adequate SNR. By adopting this approach, we expect an improvement of 3-4 stellar magnitudes in the faintest satellites that may be characterized independent of the telescope and observing waveband. Furthermore, a model derived from observations by one sensor may be used as the basis for the restoration of data sets from widely disparate telescopes and sensor modalities; data fusion in this sense is a natural feature of the approach.