FAST

Far-field Amplitude-only Speckle Transfer (FAST) algorithm for quantitative phase imaging through a multi-core fiber bundle.

updated on 14 March 2023

Quantitative phase imaging through a lensless multicore fiber endoscope

Quantitative phase imaging through a bare multicore fiber. (a) Experimental setup. SMF, single-mode fiber; MO, microscope objective; L, achromatic lens; LP, linear polarizer. (b) Digital refocusing of a multi-layer target at different axial distances. (c) 3D reconstruction of a phase target with nanoscale structures in axial direction. (d) Reconstructed optical path difference map of PAAm micro-gel beads. (e) Reconstructed optical path difference map of a HeLa cell. Scale bars 30 μm.

Table of Contents

• General Info
• Sample data
• How to use
• Setup
• Contact

General Information

• The Matlab code is ulilized for quantitative phase imaging through a multi-core fiber bundle.
• The far-field speckle images are used to retireve the phase information on the facet of the fiber bundle.
• You can read our paper for detailed information [1]. Please cite our paper if you used this code for your project.
• [1] Sun J, Wu J, Wu S, et al. Quantitative phase imaging through an ultra-thin lensless fiber endoscope[J]. Light: Science & Applications, 2022..
• https://doi.org/10.1038/s41377-022-00898-2

Sample Data

• A sample dataset can be downloaded from the link below
• https://cloudstore.zih.tu-dresden.de/index.php/s/yQq4FsWN6bFaZRQ
• The dataset contains the system response (speckle images) of a phase target
• Two far-field speckles are used to reconstruct the intrinsic phase distortion of the fiber bundle
• Another far-field speckle measured with the sample is used to reconstruct the quantitative phase information of the sample

How to use

• Requirements: Matlab 2017b or later version and at least one Nvidia GPU
• How to run the code?

1. Download and unzip the code
2. Download the sample data from the link above
3. Copy the "data.mat" to the folder where you store the code or the matlab path
4. Run the "main.m"

• Note: the phase reconstruction for the intrinsic phase distortion can take a very long time on a normal laptop, plase be patient.

Setup

• "Main.m" Main code for the reconstruction
• "prop.m" Numerical propagation function
• "tiltPhase" Generate a tilted phase image for wavefront correction

Contact

Created by Jiawei Sun (jiawei.sun@tu-dresden.de) from TU Dresden - feel free to contact me!