Nonlocal Imaging by Conditional Averaging of Random Reference Measurements

Abstract

We report the nonlocal imaging of an object by conditional averaging of the\nrandom exposure frames of a reference detector, which only sees the freely\npropagating field from a thermal light source. A bucket detector, synchronized\nwith the reference detector, records the intensity fluctuations of an identical\nbeam passing through the object mask. These fluctuations are sorted according\nto their values relative to the mean, then the reference data in the\ncorresponding time-bins for a given fluctuation range are averaged, to produce\neither positive or negative images. Since no correlation calculations are\ninvolved, this correspondence imaging technique challenges our former\ninterpretations of "ghost" imaging. Compared with conventional correlation\nimaging or compressed sensing schemes, both the number of exposures and\ncomputation time are greatly reduced, while the visibility is much improved. A\nsimple statistical model is presented to explain the phenomenon.\n