Single shot scene recovery via the bispectrum

Speckle imaging has long been used for recovery images of static scenes distorted by atmospheric optical turbulence. One limitation of speckle imaging is the need for multiple image frames where the scene is static and corrupted by independent turbulence realizations. This is a critical limitation when observing scenes containing moving objects. In this work, I use view-point images from simulated light-fields as inputs to a bispectrum-based speckle imaging algorithm. I find that near diffraction-limited imagery can be recovered from a single light-field exposure