Image-to-image translation for improved digital holographic reconstruction based on a generative adversarial network learning framework

•Based on a robust deep learning model, a generative adversarial network digital holographic reconstruction scheme is proposed for the complex and redundant holographic reconstruction difficulty.•The digital holographic image experiment acquisition system is built for the creation of image datasets suitable for the detection of targets with different optical properties.•Three-dimensional information was obtained from a sample of holographic data sets for various detection targets.