IRAN is a method of beam-combination in the hypertelescope imaging technique recently introduced by Labeyrie in optical interferometry. We propose to observe the interferometric image in the pupil plane, performing multi-axial pupil plane interferometry. Imaging is performed in a combined pupil-plane where the point-source intensity distribution (PSID) tends towards a pseudo Airy disc for a sufficiently large number of telescopes. The image is concentrated into the limited support of the output pupil of the individual telescopes, in which the object-image convolution relation is conserved. Specific deconvolution algorithms have been developped for IRAN hypertelescope imagery, based upon Lucy-like iterative techniques. We show that the classical (image plane) and IRAN (pupil plane) hypertelescope imaging techniques are equivalent if one uses optical fibers for beam transportation. An application to the VLT/VIDA concept is presented.