The Thresher : Lucky Imaging without the Waste

In traditional lucky imaging (TLI), many consecutive images of the same scene are taken with a high frame-rate camera, and all but sharpest discarded before constructing final shift-and-add image. Here we present an alternative image analysis pipeline -- The Thresher for these kinds data, based on online multi-frame blind deconvolution. It makes use available data to obtain best estimate astronomical in context reasonable computational limits; it does not require prior estimates point-spread functions images, or knowledge point sources that could provide such estimates. Most importantly, aims return is optimum justified scalar objective likelihood function. Because uses full set stack, outperforms TLI signal-to-noise; as accounts individual-frame PSFs, this without loss angular resolution. We demonstrate effectiveness our algorithm both simulated real Electron-Multiplying CCD obtained at Danish 1.54m telescope (hosted by ESO, La Silla). also explore current limitations algorithm, find choice model presented here, non-linearities flux introduced into returned scene. Ongoing development software can be viewed https://github.com/jah1994/TheThresher.