COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL

Tingyou Li; Zixin Xu; Yong S. Chu; Xiaojing Huang; Jizhou Li

doi:10.1109/ICASSP48485.2024.10447474

COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL

Tingyou Li, Zixin Xu, Yong S. Chu, Xiaojing Huang, Jizhou Li

Department of Data Science

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

2 Citations (Scopus)

Abstract

Fourier phase retrieval is essential for high-definition imaging of nanoscale structures across diverse fields, notably coherent diffraction imaging. This study presents the Single impliCit neurAl Network (SCAN), a tool built upon coordinate neural networks meticulously designed for enhanced phase retrieval performance. Remedying the drawbacks of conventional iterative methods which are easiliy trapped into local minimum solutions and sensitive to noise, SCAN adeptly connects object coordinates to their amplitude and phase within a unified network in an unsupervised manner. While many existing methods primarily use Fourier magnitude in their loss function, our approach incorporates both the predicted magnitude and phase, enhancing retrieval accuracy. Comprehensive tests validate SCAN's superiority over traditional and other deep learning models regarding accuracy and noise robustness. We also demonstrate that SCAN excels in the ptychography setting. © 2024 IEEE.

Original language	English
Title of host publication	2024 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings
Publisher	IEEE
Pages	2585-2589
ISBN (Electronic)	979-8-3503-4485-1
ISBN (Print)	979-8-3503-4486-8
DOIs	https://doi.org/10.1109/ICASSP48485.2024.10447474
Publication status	Published - 2024
Event	49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2024) - COEX, Seoul, Korea, Republic of Duration: 14 Apr 2024 → 19 Apr 2024 https://2024.ieeeicassp.org/

Publication series

Name	ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
ISSN (Print)	1520-6149
ISSN (Electronic)	2379-190X

Conference

Conference	49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2024)
Place	Korea, Republic of
City	Seoul
Period	14/04/24 → 19/04/24
Internet address	https://2024.ieeeicassp.org/

Funding

This work is supported by City University of Hong Kong under grant 9610619

Research Keywords

coherent diffraction imaging
coordinate-based neural network
Fourier phase retrieval
implicit neural representation
ptychography

Access Output

10.1109/ICASSP48485.2024.10447474

Other Access Options

Check CityUHK Library

Permanent Link

Right click to copy link

Cite this

Li, T., Xu, Z., Chu, Y. S., Huang, X., & Li, J. (2024). COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL. In 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings (pp. 2585-2589). (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). IEEE. https://doi.org/10.1109/ICASSP48485.2024.10447474

@inproceedings{59db169b290f4714981564e0aa859a21,

title = "COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL",

abstract = "Fourier phase retrieval is essential for high-definition imaging of nanoscale structures across diverse fields, notably coherent diffraction imaging. This study presents the Single impliCit neurAl Network (SCAN), a tool built upon coordinate neural networks meticulously designed for enhanced phase retrieval performance. Remedying the drawbacks of conventional iterative methods which are easiliy trapped into local minimum solutions and sensitive to noise, SCAN adeptly connects object coordinates to their amplitude and phase within a unified network in an unsupervised manner. While many existing methods primarily use Fourier magnitude in their loss function, our approach incorporates both the predicted magnitude and phase, enhancing retrieval accuracy. Comprehensive tests validate SCAN's superiority over traditional and other deep learning models regarding accuracy and noise robustness. We also demonstrate that SCAN excels in the ptychography setting. {\textcopyright} 2024 IEEE.",

keywords = "coherent diffraction imaging, coordinate-based neural network, Fourier phase retrieval, implicit neural representation, ptychography",

author = "Tingyou Li and Zixin Xu and Chu, {Yong S.} and Xiaojing Huang and Jizhou Li",

year = "2024",

doi = "10.1109/ICASSP48485.2024.10447474",

language = "English",

isbn = "979-8-3503-4486-8",

series = "ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings",

publisher = "IEEE",

pages = "2585--2589",

booktitle = "2024 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings",

address = "United States",

note = "49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2024) ; Conference date: 14-04-2024 Through 19-04-2024",

url = "https://2024.ieeeicassp.org/",

}

Li, T, Xu, Z, Chu, YS, Huang, X & Li, J 2024, COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL. in 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings. ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings, IEEE, pp. 2585-2589, 49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2024), Seoul, Korea, Republic of, 14/04/24. https://doi.org/10.1109/ICASSP48485.2024.10447474

COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL. / Li, Tingyou; Xu, Zixin; Chu, Yong S. et al.
2024 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings. IEEE, 2024. p. 2585-2589 (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings).

Research output: Chapters, Conference Papers, Creative and Literary Works › RGC 32 - Refereed conference paper (with host publication) › peer-review

TY - GEN

T1 - COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL

AU - Li, Tingyou

AU - Xu, Zixin

AU - Chu, Yong S.

AU - Huang, Xiaojing

AU - Li, Jizhou

PY - 2024

Y1 - 2024

N2 - Fourier phase retrieval is essential for high-definition imaging of nanoscale structures across diverse fields, notably coherent diffraction imaging. This study presents the Single impliCit neurAl Network (SCAN), a tool built upon coordinate neural networks meticulously designed for enhanced phase retrieval performance. Remedying the drawbacks of conventional iterative methods which are easiliy trapped into local minimum solutions and sensitive to noise, SCAN adeptly connects object coordinates to their amplitude and phase within a unified network in an unsupervised manner. While many existing methods primarily use Fourier magnitude in their loss function, our approach incorporates both the predicted magnitude and phase, enhancing retrieval accuracy. Comprehensive tests validate SCAN's superiority over traditional and other deep learning models regarding accuracy and noise robustness. We also demonstrate that SCAN excels in the ptychography setting. © 2024 IEEE.

AB - Fourier phase retrieval is essential for high-definition imaging of nanoscale structures across diverse fields, notably coherent diffraction imaging. This study presents the Single impliCit neurAl Network (SCAN), a tool built upon coordinate neural networks meticulously designed for enhanced phase retrieval performance. Remedying the drawbacks of conventional iterative methods which are easiliy trapped into local minimum solutions and sensitive to noise, SCAN adeptly connects object coordinates to their amplitude and phase within a unified network in an unsupervised manner. While many existing methods primarily use Fourier magnitude in their loss function, our approach incorporates both the predicted magnitude and phase, enhancing retrieval accuracy. Comprehensive tests validate SCAN's superiority over traditional and other deep learning models regarding accuracy and noise robustness. We also demonstrate that SCAN excels in the ptychography setting. © 2024 IEEE.

KW - coherent diffraction imaging

KW - coordinate-based neural network

KW - Fourier phase retrieval

KW - implicit neural representation

KW - ptychography

UR - http://www.scopus.com/inward/record.url?scp=85195368518&partnerID=8YFLogxK

UR - https://www.scopus.com/record/pubmetrics.uri?eid=2-s2.0-85195368518&origin=recordpage

U2 - 10.1109/ICASSP48485.2024.10447474

DO - 10.1109/ICASSP48485.2024.10447474

M3 - RGC 32 - Refereed conference paper (with host publication)

SN - 979-8-3503-4486-8

T3 - ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings

SP - 2585

EP - 2589

BT - 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings

PB - IEEE

T2 - 49th IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP 2024)

Y2 - 14 April 2024 through 19 April 2024

ER -

Li T, Xu Z, Chu YS, Huang X, Li J. COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL. In 2024 IEEE International Conference on Acoustics, Speech, and Signal Processing - Proceedings. IEEE. 2024. p. 2585-2589. (ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings). Epub 2024 Mar 18. doi: 10.1109/ICASSP48485.2024.10447474

COORDINATE-BASED NEURAL NETWORK FOR FOURIER PHASE RETRIEVAL

Abstract

Publication series

Conference

Funding

Research Keywords

Access Output

Other Access Options

Permanent Link

Other Files and Links

Fingerprint

Cite this