OAH-Net is a novel deep neural network architecture that leverages the physical principles of off-axis holography and weakly supervised learning to achieve real-time, accurate, and generalizable hologram reconstruction for high-throughput digital holographic microscopy applications.
The NU-FDS algorithm enables fast and accurate reconstruction of ultra-wide-angle computer-generated holograms (CGHs) by using non-uniform frequency magnification to correct the axial distance of parabolic waves, allowing for the application of the Fresnel Transform (FrT) method.