On the use of machine learning for solving computational imaging problems
It has recently been recognized that compressed sensing, especially dictionaries and related methods, formally map to machine learning architectures, e.g. recurrent neural networks. This has led to rapid growth in algorithms and methods based on deep neural networks (but not only) for solving a vari...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
SPIE
2021
|
| Online Access: | https://hdl.handle.net/1721.1/129723 |
| _version_ | 1811076744046706688 |
|---|---|
| author | Barbastathis, George |
| author2 | Massachusetts Institute of Technology. Department of Mechanical Engineering |
| author_facet | Massachusetts Institute of Technology. Department of Mechanical Engineering Barbastathis, George |
| author_sort | Barbastathis, George |
| collection | MIT |
| description | It has recently been recognized that compressed sensing, especially dictionaries and related methods, formally map to machine learning architectures, e.g. recurrent neural networks. This has led to rapid growth in algorithms and methods based on deep neural networks (but not only) for solving a variety of inverse and computational imaging problems. In this paper, we review these developments in the specific context of quantitative phase imaging and emphasizing the impact of object power spectral density and noise properties on the quality of the reconstructions. |
| first_indexed | 2024-09-23T10:26:45Z |
| format | Article |
| id | mit-1721.1/129723 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T10:26:45Z |
| publishDate | 2021 |
| publisher | SPIE |
| record_format | dspace |
| spelling | mit-1721.1/1297232022-09-26T17:57:11Z On the use of machine learning for solving computational imaging problems Barbastathis, George Massachusetts Institute of Technology. Department of Mechanical Engineering Singapore-MIT Alliance in Research and Technology (SMART) It has recently been recognized that compressed sensing, especially dictionaries and related methods, formally map to machine learning architectures, e.g. recurrent neural networks. This has led to rapid growth in algorithms and methods based on deep neural networks (but not only) for solving a variety of inverse and computational imaging problems. In this paper, we review these developments in the specific context of quantitative phase imaging and emphasizing the impact of object power spectral density and noise properties on the quality of the reconstructions. 2021-02-09T19:32:02Z 2021-02-09T19:32:02Z 2020-02 2020-06-22T19:01:01Z Article http://purl.org/eprint/type/ConferencePaper 9781510632615 9781510632622 1996-756X https://hdl.handle.net/1721.1/129723 Barbastathis, George. "On the use of machine learning for solving computational imaging problems." Proceedings of SPIE (February 2020) © 2020 SPIE. en 10.1117/12.2554397 Proceedings of SPIE Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use. application/pdf SPIE SPIE |
| spellingShingle | Barbastathis, George On the use of machine learning for solving computational imaging problems |
| title | On the use of machine learning for solving computational imaging problems |
| title_full | On the use of machine learning for solving computational imaging problems |
| title_fullStr | On the use of machine learning for solving computational imaging problems |
| title_full_unstemmed | On the use of machine learning for solving computational imaging problems |
| title_short | On the use of machine learning for solving computational imaging problems |
| title_sort | on the use of machine learning for solving computational imaging problems |
| url | https://hdl.handle.net/1721.1/129723 |
| work_keys_str_mv | AT barbastathisgeorge ontheuseofmachinelearningforsolvingcomputationalimagingproblems |