Deep learning-empowered wavefront shaping in scattering media
Wavefront shaping is a widely accepted approach to focus light within or through scattering media, however, so far, most implementations to pre-compensate the optical wavefronts have only operated with static media due to the requirements of iterative optimizations or measurement of the transmission...
Saved in:
Main Author: | |
---|---|
Other Authors: | |
Format: | Thesis-Doctor of Philosophy |
Language: | English |
Published: |
Nanyang Technological University
2021
|
Subjects: | |
Online Access: | https://hdl.handle.net/10356/147459 |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Nanyang Technological University |
Language: | English |
Summary: | Wavefront shaping is a widely accepted approach to focus light within or through scattering media, however, so far, most implementations to pre-compensate the optical wavefronts have only operated with static media due to the requirements of iterative optimizations or measurement of the transmission matrix, which are time-consuming. With the goal to comprehensively resolve wavefront shaping problems through nonstationary scattering media, this Ph.D. thesis comprehensively investigates the fundamental physics of scattering and inverse scattering in disordered media. A reinforced hybrid algorithm is proposed to improve wavefront shaping efficiency. Moreover, deep learning frameworks are developed based on the mathematical models, and light focusing and focusing recovery through scattering media with perturbations, media that are continually changing at constant speeds, and randomly altering media are all achieved. |
---|