Multitarget scattering imaging based on Nonnegative Matrix Factorization beyond the optical memory effect
Speckle correlation technique is limited by the optical memory effect (OME) range, and aliasing phenomenon occurs in multitarget information beyond the range. Here, we propose a multitarget scattering imaging method based on Nonnegative Matrix Factorization (NMF) beyond the OME, which can separate a...
Gespeichert in:
Veröffentlicht in: | Optics communications 2024-05, Vol.558, p.130350, Article 130350 |
---|---|
Hauptverfasser: | , , , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | Speckle correlation technique is limited by the optical memory effect (OME) range, and aliasing phenomenon occurs in multitarget information beyond the range. Here, we propose a multitarget scattering imaging method based on Nonnegative Matrix Factorization (NMF) beyond the OME, which can separate and interpret the single target information from the aliasing information. Based on the study of the multitarget aliasing speckle autocorrelation model, we apply NMF to unmix the multitarget speckle autocorrelation beyond the OME, and combine with phase retrieval (PR) technology to recover the target. Our approach requires only single-shot speckle information without prior information to realize wide-field multitarget imaging. This work provides potential value for biomedical imaging and astronomical observation, and helps to broaden the application scope of speckle correlation technique.
•This method can solve the problem of speckle information aliasing in multitarget scattering imaging without prior information.•This method can achieve scattering imaging of multiple targets with a spacing of 15 mm, which expands the range of optical memory effect by 5 times.•The experimental setup is simple and feasible, requiring only a single-frame speckle. |
---|---|
ISSN: | 0030-4018 1873-0310 |
DOI: | 10.1016/j.optcom.2024.130350 |