The predicted sharp cutoff of the transfer function of a coherent imaging system is not seen experimentally. The experiment suggests that this is due to the neglect of the diffraction, caused by the finite lens, of the individual plane-wave components of the angular spectrum of the object wave. It is shown that this omission is equivalent to dropping a quadratic phase term in the description of the image obtained with the Fresnel-Kirchhoff diffraction formula.
W. T. Cathey, "Angular spectrum and coherent imaging," J. Opt. Soc. Am. 64, 1503-1506 (1974)