Single-pixel edge enhancement of object via convolutional filtering with localized vortex phase

Abstract

In this paper, we propose and numerically demonstrate an all-directional edge-enhanced microscopy based on single-pixel imaging using convolutional filtering with a localized vortex phase. The edge-enhanced microscope with a vortex filter is of particular interest for optical information processing, as it can highlight both amplitude and the phase edges of complex objects in all directions. While Fourier-domain single-pixel imaging has been proposed as a cost-effective approach for edge enhancement in non-visible wavelengths, it requires three or four times more measurements due to the need for three- or four-phase shifting to detect optical complex amplitudes in the Fourier domain. Our method implements convolutional filtering with a localized vortex phase in the spatial domain, eliminating the need for these extra measurements required by phase-shifting methods. Simulations demonstrate a high correlation coefficient of 0.95 between ideal and enhanced edges, offering significant improvements for edge enhancement in various invisible-wavelength imaging applications.