Sequential Image Recovery Using Joint Hierarchical Bayesian Learning
Abstract Recovering temporal image sequences (videos) based on indirect, noisy, or incomplete data is an essential yet challenging task. We specifically consider the case where each data set is missing vital information, which prevents the accurate recovery of the individual images. A...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Article |
| Language: | English |
| Published: |
Springer US
2023
|
| Online Access: | https://hdl.handle.net/1721.1/150788 |
| _version_ | 1826216106570809344 |
|---|---|
| author | Xiao, Yao Glaubitz, Jan |
| author2 | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics |
| author_facet | Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Xiao, Yao Glaubitz, Jan |
| author_sort | Xiao, Yao |
| collection | MIT |
| description | Abstract
Recovering temporal image sequences (videos) based on indirect, noisy, or incomplete data is an essential yet challenging task. We specifically consider the case where each data set is missing vital information, which prevents the accurate recovery of the individual images. Although some recent (variational) methods have demonstrated high-resolution image recovery based on jointly recovering sequential images, there remain robustness issues due to parameter tuning and restrictions on the type of sequential images. Here, we present a method based on hierarchical Bayesian learning for the joint recovery of sequential images that incorporates prior intra- and inter-image information. Our method restores the missing information in each image by “borrowing” it from the other images. More precisely, we couple sequential images by penalizing their pixel-wise difference. The corresponding penalty terms (one for each pixel and pair of subsequent images) are treated as weakly-informative random variables that favor small pixel-wise differences but allow occasional outliers. As a result, all of the individual reconstructions yield improved accuracy. Our method can be used for various data acquisitions and allows for uncertainty quantification. Some preliminary results indicate its potential use for sequential deblurring and magnetic resonance imaging. |
| first_indexed | 2024-09-23T16:42:24Z |
| format | Article |
| id | mit-1721.1/150788 |
| institution | Massachusetts Institute of Technology |
| language | English |
| last_indexed | 2024-09-23T16:42:24Z |
| publishDate | 2023 |
| publisher | Springer US |
| record_format | dspace |
| spelling | mit-1721.1/1507882024-01-08T20:47:49Z Sequential Image Recovery Using Joint Hierarchical Bayesian Learning Xiao, Yao Glaubitz, Jan Massachusetts Institute of Technology. Department of Aeronautics and Astronautics Abstract Recovering temporal image sequences (videos) based on indirect, noisy, or incomplete data is an essential yet challenging task. We specifically consider the case where each data set is missing vital information, which prevents the accurate recovery of the individual images. Although some recent (variational) methods have demonstrated high-resolution image recovery based on jointly recovering sequential images, there remain robustness issues due to parameter tuning and restrictions on the type of sequential images. Here, we present a method based on hierarchical Bayesian learning for the joint recovery of sequential images that incorporates prior intra- and inter-image information. Our method restores the missing information in each image by “borrowing” it from the other images. More precisely, we couple sequential images by penalizing their pixel-wise difference. The corresponding penalty terms (one for each pixel and pair of subsequent images) are treated as weakly-informative random variables that favor small pixel-wise differences but allow occasional outliers. As a result, all of the individual reconstructions yield improved accuracy. Our method can be used for various data acquisitions and allows for uncertainty quantification. Some preliminary results indicate its potential use for sequential deblurring and magnetic resonance imaging. 2023-05-22T13:57:05Z 2023-05-22T13:57:05Z 2023-05-18 2023-05-21T03:12:06Z Article http://purl.org/eprint/type/JournalArticle https://hdl.handle.net/1721.1/150788 Journal of Scientific Computing. 2023 May 18;96(1):4 PUBLISHER_CC en https://doi.org/10.1007/s10915-023-02234-1 Creative Commons Attribution http://creativecommons.org/licenses/by/4.0/ The Author(s) application/pdf Springer US Springer US |
| spellingShingle | Xiao, Yao Glaubitz, Jan Sequential Image Recovery Using Joint Hierarchical Bayesian Learning |
| title | Sequential Image Recovery Using Joint Hierarchical Bayesian Learning |
| title_full | Sequential Image Recovery Using Joint Hierarchical Bayesian Learning |
| title_fullStr | Sequential Image Recovery Using Joint Hierarchical Bayesian Learning |
| title_full_unstemmed | Sequential Image Recovery Using Joint Hierarchical Bayesian Learning |
| title_short | Sequential Image Recovery Using Joint Hierarchical Bayesian Learning |
| title_sort | sequential image recovery using joint hierarchical bayesian learning |
| url | https://hdl.handle.net/1721.1/150788 |
| work_keys_str_mv | AT xiaoyao sequentialimagerecoveryusingjointhierarchicalbayesianlearning AT glaubitzjan sequentialimagerecoveryusingjointhierarchicalbayesianlearning |