High-resolution ghost image and ghost diffraction experiments are performed by using a single classical source of pseudothermal speckle light divided by a beam splitter. Passing from the image to the diffraction result solely relies on changing the optical setup in the reference arm, while leaving the object arm untouched. The product of spatial resolutions of the ghost image and ghost diffraction experiments is shown to overcome a limit which seemed to be achievable only with entangled photons.

We present a new technique, differential ghost imaging (DGI), which dramatically enhances the signal-to-noise ratio (SNR) of imaging methods based on spatially correlated beams. DGI can measure the transmission function of an object in absolute units, with a SNR that can be orders of magnitude higher than the one achievable with the conventional ghost imaging (GI) analysis. This feature allows for the first time, to our knowledge, the imaging of weakly absorbing objects, which represents a breakthrough for GI applications. Theoretical analysis and experimental and numerical data assessing the performances of the technique are presented.

We investigate experimentally fundamental properties of coherent ghost imaging using spatially incoherent beams generated from a pseudo-thermal source. A complementarity between the coherence of the beams and the correlation between them is demonstrated by showing a complementarity between ghost diffraction and ordinary diffraction patterns. In order for the ghost imaging scheme to work it is therefore crucial to have incoherent beams. The visibility of the information is shown for the ghost image to become better as the object size relative to the speckle size is decreased, and therefore a remarkable tradeoff between resolution and visibility exists. The experimental conclusions are backed up by both theory and numerical simulations.

We present a PC-based multi-tau software correlator suitable for processing dynamic light-scattering data. The correlator is based on a simple algorithm that was developed with the graphical programming language LabVIEW, according to which the incoming data are processed on line without any storage on the hard disk. By use of a standard photon-counting unit, a National Instruments Model 6602-PCI timer-counter, and a 550-MHz Pentium III personal computer, correlation functions can be worked out in full real-time over time scales of ~5 mus and in batch processing down to time scales of ~300 ns. The latter limit is imposed by the speed of data transfer between the counter and the PC's memory and thus is prone to be progressively reduced with future technological development. Testing of the correlator and evaluation of its performances were carried out by use of dilute solutions of calibrated polystyrene spheres. Our results indicate that the correlation functions are determined with such precision that the corresponding particle diameters can be recovered to within an accuracy of a few percent rms.