Single-shot complex amplitude demodulation based on deep learning

Abstract

This paper proposes a complex amplitude demodulation method based on deep learning used in holographic data storage (HDS). To increase the storage capacity of a single data page in HDS, the complex amplitude of the object light can be used to encode the information data. However, the phase information of the complex amplitude cannot be detected directly. In this paper, we propose a non-interferometric complex amplitude retrieval method based on deep learning that can demodulate amplitude and phase simultaneously. A one-to-two convolutional neural network (CNN) is designed to establish the relationship between the intensity images captured by the detector and complex amplitude data pages. A simulation experiment is established to verify the feasibility of the proposed method.