Emergent physics-informed design of deep learning for microscopy

Research output: Contribution to journalReview articlepeer-review

9 Citations (Scopus)
9 Downloads (Pure)

Abstract

Deep learning has revolutionised microscopy, enabling automated means for image classification, tracking and transformation. Going beyond machine vision, deep learning has recently emerged as a universal and powerful tool to address challenging and previously untractable image recovery problems. In seeking accurate, learned means of inversion, these advances have transformed conventional deep learning methods to those cognisant of the underlying physics of image formation, enabling robust, efficient and accurate recovery even in severely ill-posed conditions. In this Perspective, we explore the emergence of physics-informed deep learning that will enable universal and accessible computational microscopy.
Original languageEnglish
Article number021003
Number of pages11
JournalJournal of Physics: Photonics
Volume3
Issue number2
Early online date14 Apr 2021
DOIs
Publication statusPublished - Apr 2021

Keywords

  • Deep learning
  • Microscopy
  • Inverse methods
  • Physics-informed learning
  • Computational imaging

Fingerprint

Dive into the research topics of 'Emergent physics-informed design of deep learning for microscopy'. Together they form a unique fingerprint.

Cite this