An optimized scheme of computer-generated hologram compression with convolutional neural networks and genetic algorithm

Holography is pivotal in various applications ranging from microscopy to data storage. However, the practical implementation of holography often necessitates compression, which inevitably degrades the quality of the reconstructed images. Addressing this challenge, a scheme of computer-generated hologram compression with convolutional neural networks optimized is proposed. In this scheme, our research investigates the potential of genetic algorithm, a class of evolutionary algorithms inspired by natural selection, to optimize the initial weights of the network. Experimental results show that the scheme is feasible.