Bayesian sparse reconstruction based on dictionary learning

Yan Wang; Jun Ke

doi:10.1117/12.2575180

Bayesian sparse reconstruction based on dictionary learning

Yan Wang^*, Jun Ke

^*Corresponding author for this work

School of Optics and Photonics

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

Imaging through thick scattering media produces a random speckle signal with wealth information, which can be restored by subsequent processing. While a moving target is hard to reconstruct by existing technology, we apply temporal Bayesian compressed sensing method to overcome this limitation. In addition, an over completed dictionary is used as a sparse base to improve the accuracy of the reconstructions. In this letter, we improve system time resolution without changing its spatial resolution and reconstruct T frame speckle images from a single temporal compressed speckle measurement.

Original language	English
Title of host publication	Advanced Optical Imaging Technologies III
Editors	Xiao-Cong Yuan, P. Scott Carney, Kebin Shi
Publisher	SPIE
ISBN (Electronic)	9781510639133
DOIs	https://doi.org/10.1117/12.2575180
Publication status	Published - 2020
Event	Advanced Optical Imaging Technologies III 2020 - Virtual, Online, China Duration: 11 Oct 2020 → 16 Oct 2020

Publication series

Name	Proceedings of SPIE - The International Society for Optical Engineering
Volume	11549
ISSN (Print)	0277-786X
ISSN (Electronic)	1996-756X

Conference

Conference	Advanced Optical Imaging Technologies III 2020
Country/Territory	China
City	Virtual, Online
Period	11/10/20 → 16/10/20

Keywords

Bayesian Estimation
Compressed Sensing
Dictionary Learning

Access to Document

10.1117/12.2575180

Cite this

Wang, Y., & Ke, J. (2020). Bayesian sparse reconstruction based on dictionary learning. In X.-C. Yuan, P. S. Carney, & K. Shi (Eds.), Advanced Optical Imaging Technologies III Article 115491L (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11549). SPIE. https://doi.org/10.1117/12.2575180

@inproceedings{ad7ff385dc5b4bc2b39388344186fb6e,

title = "Bayesian sparse reconstruction based on dictionary learning",

abstract = "Imaging through thick scattering media produces a random speckle signal with wealth information, which can be restored by subsequent processing. While a moving target is hard to reconstruct by existing technology, we apply temporal Bayesian compressed sensing method to overcome this limitation. In addition, an over completed dictionary is used as a sparse base to improve the accuracy of the reconstructions. In this letter, we improve system time resolution without changing its spatial resolution and reconstruct T frame speckle images from a single temporal compressed speckle measurement.",

keywords = "Bayesian Estimation, Compressed Sensing, Dictionary Learning",

author = "Yan Wang and Jun Ke",

note = "Publisher Copyright: {\textcopyright} 2020 SPIE; Advanced Optical Imaging Technologies III 2020 ; Conference date: 11-10-2020 Through 16-10-2020",

year = "2020",

doi = "10.1117/12.2575180",

language = "English",

series = "Proceedings of SPIE - The International Society for Optical Engineering",

publisher = "SPIE",

editor = "Xiao-Cong Yuan and Carney, {P. Scott} and Kebin Shi",

booktitle = "Advanced Optical Imaging Technologies III",

address = "United States",

}

Wang, Y & Ke, J 2020, Bayesian sparse reconstruction based on dictionary learning. in X-C Yuan, PS Carney & K Shi (eds), Advanced Optical Imaging Technologies III., 115491L, Proceedings of SPIE - The International Society for Optical Engineering, vol. 11549, SPIE, Advanced Optical Imaging Technologies III 2020, Virtual, Online, China, 11/10/20. https://doi.org/10.1117/12.2575180

Bayesian sparse reconstruction based on dictionary learning. / Wang, Yan; Ke, Jun.
Advanced Optical Imaging Technologies III. ed. / Xiao-Cong Yuan; P. Scott Carney; Kebin Shi. SPIE, 2020. 115491L (Proceedings of SPIE - The International Society for Optical Engineering; Vol. 11549).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - Bayesian sparse reconstruction based on dictionary learning

AU - Wang, Yan

AU - Ke, Jun

PY - 2020

Y1 - 2020

N2 - Imaging through thick scattering media produces a random speckle signal with wealth information, which can be restored by subsequent processing. While a moving target is hard to reconstruct by existing technology, we apply temporal Bayesian compressed sensing method to overcome this limitation. In addition, an over completed dictionary is used as a sparse base to improve the accuracy of the reconstructions. In this letter, we improve system time resolution without changing its spatial resolution and reconstruct T frame speckle images from a single temporal compressed speckle measurement.

AB - Imaging through thick scattering media produces a random speckle signal with wealth information, which can be restored by subsequent processing. While a moving target is hard to reconstruct by existing technology, we apply temporal Bayesian compressed sensing method to overcome this limitation. In addition, an over completed dictionary is used as a sparse base to improve the accuracy of the reconstructions. In this letter, we improve system time resolution without changing its spatial resolution and reconstruct T frame speckle images from a single temporal compressed speckle measurement.

KW - Bayesian Estimation

KW - Compressed Sensing

KW - Dictionary Learning

UR - http://www.scopus.com/inward/record.url?scp=85097170041&partnerID=8YFLogxK

U2 - 10.1117/12.2575180

DO - 10.1117/12.2575180

M3 - Conference contribution

AN - SCOPUS:85097170041

T3 - Proceedings of SPIE - The International Society for Optical Engineering

BT - Advanced Optical Imaging Technologies III

A2 - Yuan, Xiao-Cong

A2 - Carney, P. Scott

A2 - Shi, Kebin

PB - SPIE

T2 - Advanced Optical Imaging Technologies III 2020

Y2 - 11 October 2020 through 16 October 2020

ER -

Bayesian sparse reconstruction based on dictionary learning

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this