论文信息 - Enhancing resolution in coherent microscopy using deep learning

Enhancing resolution in coherent microscopy using deep learning

A generative adversarial network (GAN) based super-resolution framework is presented. This deep learning-based framework is capable of enhancing the resolution of coherent imaging systems in both pixel size-limited and diffraction-limited microscopy systems. © 2019 The Author(s)

[1] Yibo Zhang,et al. Wide-field computational imaging of pathology slides using lens-free on-chip microscopy , 2014, Science Translational Medicine.

[2] Lei Tian,et al. Compressive holographic two-dimensional localization with 1/30(2) subpixel accuracy. , 2014, Optics express.

[3] Zeev Zalevsky,et al. Superresolved imaging in digital holography by superposition of tilted wavefronts. , 2006, Applied optics.

[4] Yibo Zhang,et al. Deep Learning Microscopy , 2017, ArXiv.

[5] Yibo Zhang,et al. Deep learning-based super-resolution in coherent imaging systems , 2018, Scientific Reports.

[6] Yoshua Bengio,et al. Generative Adversarial Nets , 2014, NIPS.