Abstract
We present a deep learning framework based on a generative adversarial network (GAN) to perform super-resolution in coherent imaging systems. We demonstrate that this framework can enhance the resolution of both pixel size-limited and diffraction-limited coherent imaging systems. The capabilities of this approach are experimentally validated by super-resolving complex-valued images acquired using a lensfree on-chip holographic microscope, the resolution of which was pixel size-limited. Using the same GAN-based approach, we also improved the resolution of a lens-based holographic imaging system that was limited in resolution by the numerical aperture of its objective lens. This deep learning-based super-resolution framework can be broadly applied to enhance the space-bandwidth product of coherent imaging systems using image data and convolutional neural networks, and provides a rapid, non-iterative method for solving inverse image reconstruction or enhancement problems in optics.
Highlights
Coherent imaging systems have many advantages for applications where the specimen’s complex field information is of interest[1]
We first report the performance of the network when applied to the pixel size-limited coherent imaging system using a Pap smear sample and a Masson’s trichrome stained lung tissue section
We report the structural similarity index (SSIM) values with respect to the reference label images in order to further evaluate the performance of our network output when applied to a pixel size-limited coherent imaging system
Summary
Coherent imaging systems have many advantages for applications where the specimen’s complex field information is of interest[1]. Deep-learning based holographic image reconstruction techniques have been demonstrated to create a high-fidelity reconstruction from a single in-line hologram[11,12,13], and are capable of further extending the depth-of-field of the reconstructed image[14]. Several approaches have been demonstrated to improve the resolution of coherent imaging systems[15,16,17,18,19,20] Most of these techniques require sequential measurements and assume that the object is quasi-static while a diverse set of measurements are performed on it. We quantify our results using the structural similarity index (SSIM)[32] and spatial frequency content of the network’s output images in comparison to the higher resolution images (which constitute our ground truth) This data-driven image super-resolution framework is applicable to enhance the performance of various coherent imaging systems
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
