Comment on "Can Two-Photon Correlation of Chaotic Light Be Considered as Correlation of Intensity Fluctuations?".

A recent Letter [1] presents an experiment of ghost imaging with chaotic light. We definitely disagree with two main claims of the accompanying theory. The first one is ‘‘the explanation in terms of statistical correlation of intensity fluctuations would not give an acceptable interpretation for this experiment.’’ In the experiment of [1] (analogous to the ghost image experiment of [2]), a beam of chaotic light from a pseudothermal source is divided by a beam splitter (BS). Beam 1 probes an object at distance dA from the source, and the light is collected into a bucket detector. Beam 2 is detected at a distance dB dA from the source. Two-photon coincidences are registered, whose rate is proportional to G2 x1;x2 hI1 x1 I2 x2 i, Ii being the intensities of fields Ei and h i a statistical average. By using the (classical) Siegert relation for chaotic statistics and the BS transformation, we find (see, e.g., [3])