Stellar object image recovery using neural network atmospheric modelling

This paper describes the use of a back-propagation neural network to model a turbulent atmosphere. The modeling of the atmosphere allows the near real time resolution of a stellar object's image from its respective speckle images. These speckle images are produced when the light from a stellar object passes through a diffusing medium such as our atmosphere. If this modeling is eventually successful it will provide the astronomer the capability of near real time stellar object identification at low cost and with fairly simple technology.