Computational ghost imaging with the sweeping algorithm

Abstract

The idea of ghost imaging was firstly presented more than 25 years ago and since then many techniques have been proposed to improve image quality and simplify optical scheme. Nevertheless, the image reconstruction speed has always remained a serious problem to wide practical implementations. As a solution, a sweeping algorithm was proposed, which allowed to reduce the number of iterations to image resolution. However, this method has some serious limitations on objects shape. In this article we present a new technique, which combine a sweeping algorithm using alternating rows/columns and traditional pseudothermal ghost imaging using speckle structures.