This Letter aims to propose a dynamic-range compression and decompression scheme for digital holograms that uses a deep neural network (DNN). The proposed scheme uses simple thresholding to compress the dynamic range of holograms with 8-bit gradation to binary holograms. Although this can decrease the amount of data by one-eighth, the binarization strongly degrades the image quality of the reconstructed images. The proposed scheme uses a DNN to predict the original gradation holograms from the binary holograms, and the error-diffusion algorithm of the binarization process contributes significantly to training the DNN. The performance of the scheme exceeds that of modern compression techniques such as JPEG 2000 and high-efficiency video coding.
Shimobaba, T, Blinder, D, Makowski, M, Schelkens, P, Yamamoto, Y, Hoshi, I, Nishitsuji, T, Endo, Y, Kakue, T & Ito, T 2019, 'Dynamic-range compression scheme for digital hologram using a deep neural network', Optics Letters, vol. 44, no. 12, pp. 3038-3041. https://doi.org/10.1364/OL.44.003038
Shimobaba, T., Blinder, D., Makowski, M., Schelkens, P., Yamamoto, Y., Hoshi, I., Nishitsuji, T., Endo, Y., Kakue, T., & Ito, T. (2019). Dynamic-range compression scheme for digital hologram using a deep neural network. Optics Letters, 44(12), 3038-3041. https://doi.org/10.1364/OL.44.003038
@article{15b871c39ca8481fa370d0563b33c772,
title = "Dynamic-range compression scheme for digital hologram using a deep neural network",
abstract = "This Letter aims to propose a dynamic-range compression and decompression scheme for digital holograms that uses a deep neural network (DNN). The proposed scheme uses simple thresholding to compress the dynamic range of holograms with 8-bit gradation to binary holograms. Although this can decrease the amount of data by one-eighth, the binarization strongly degrades the image quality of the reconstructed images. The proposed scheme uses a DNN to predict the original gradation holograms from the binary holograms, and the error-diffusion algorithm of the binarization process contributes significantly to training the DNN. The performance of the scheme exceeds that of modern compression techniques such as JPEG 2000 and high-efficiency video coding.",
keywords = "holography, compression, coding, deep neural network, artificial intelligence, DNN, JPEG 2000",
author = "Tomoyoshi Shimobaba and David Blinder and Michal Makowski and Peter Schelkens and Yoya Yamamoto and Ikuo Hoshi and Takashi Nishitsuji and Yutaka Endo and Takashi Kakue and Tomoyoshi Ito",
year = "2019",
month = jun,
day = "15",
doi = "10.1364/OL.44.003038",
language = "English",
volume = "44",
pages = "3038--3041",
journal = "Optics Letters",
issn = "0146-9592",
publisher = "Optical Society of America",
number = "12",
}