Single-shot phase-shifting incoherent digital holography with multiplexed checkerboard phase gratings is proposed for acquiring holograms of moving objects. The gratings presented here play the following three roles: dividing the beams, modulating the curvature of spherical beams, and introducing different phase shifts. With the gratings of our proposed method, four individual holograms of a spatially incoherent light are formed on an image sensor. Therefore, it is possible to simultaneously capture four holograms and implement a phase-shifting technique. A proof-of-principle experiment was conducted to show the feasibility of the proposed method.

In angle-multiplexed holographic memory, the full width at half maximum of the Bragg selectivity curves is dependent on the angle formed between the medium and incident laser beams. This indicates the possibility of high density and high multiplexing number by varying the angular intervals between adjacent holograms. We propose an angular interval scheduling for closely stacking holograms into medium even when the angle range is limited. We obtained bit error rates of the order of 10 À4 under the following conditions: medium thickness of 1 mm, laser beam wavelength of 532 nm, and angular multiplexing number of 300.

Photopolymer materials shrink because of photopolymerization. This shrinkage distorts the recorded interference fringes in a medium made of such material, which in turn degrades the reconstructed image quality. Adaptive optics controlled by a genetic algorithm was developed to optimize the wavefront of the reference beam while reproducing in order to compensate for the interference fringe distortion. We defined a fitness measure for this genetic algorithm that involves the mean brightness and coefficients of the variations of bit data "1" and "0". In an experiment, the adaptive optics improved the reconstructed image to the extent that data could be reproduced from the entire area of the image, and the signal to noise ratio of the reproduced data could be improved.

We propose a data demodulation method based on a deep-learning algorithm. A convolutional neural network (CNN), which can accurately classify images, was used in the demodulation of data reproduced from holographic data storage (HDS). We designed CNNs and taught them the rules for demodulation based on the optical characteristics of the HDS using 700 reproduced data pages. The CNNs that learned could demodulate the data and decrease the number of demodulation errors by about 75% compared with hard decision image classification methods. This result showed an improvement in optical noise tolerance, which enhances the HDS with larger capacity and higher data-transfer rate.

Photopolymer materials are feasible for holographic recording media. However, these materials shrink owing to photopolymerization and interference fringes recorded in them distort. In addition, temperature variation causes shrinkage and expansion of these materials and thus distorts recorded interference fringes. This distortion degrades reconstructed image quality and decreases the signal-to-noise ratio of the reproduced data. We applied adaptive optics controlled by a genetic algorithm to compensate for the distortion and improved the reconstructed image quality at 25 and 30 °C ambient temperature. Under these conditions, the signal-to-noise ratio of reproduced data was more than 4 dB. Furthermore, we evaluated the distortion due to the temperature variation by using a medium angle and the wavefront of the reference beam. We found that the distortion caused by anisotropic shrinkage is slight; consequently, an optimised wavefront at 25 °C can compensate for the interference fringe distortion and increase the signal-to-noise ratio by adjusting only the medium angle even if a temperature variation occurs. Adaptive optics can thus be used to compensate for interference fringe distortion caused by shrinkage and expansion due to temperature variation.