Temporal Super-Resolution Microscopy Using a Hue-Encoded Shutter
OSA Biomedical Optics Express Volume 10 Issue 09, September 2019
Authors of the paper:
- Christian Jaques, Idiap, EPFL: personnal page at Idiap. Do not hesitate to contact me.
- Emmanuel Pignat, Idiap, EPFL: personnal page at Idiap
- Sylvain Calinon, Idiap, EPFL: personnal page at Idiap
- Michael Liebling, Idiap, UCSB : personnal page at Idiap
This repository holds the code and instruction to reproduce the method presented in the paper entitled Temporal Super-Resolution Microscopy Using a Hue-Encoded Shutter.
There are three main parts to the repository: (1) code and instructions related to data acquisition (
acquisition folder), (2) post-acquisition processing code (
processing folder) and (3) the data acquired to make the figures in the paper (
After setting up the physical system, see
readme files within each folder for respective instructions.
These instructions are meant for Debian distributions of Linux (tested on Debian Stretch (4.9) and Ubuntu 16.04 LTS). The data processing should work on all platforms. The data acquisition should work on Windows (the wrapping of the
C++ code can be tricky), for macosX, the required
ueye library isn't available.
If someone has a way to drive a Thorcam USB3 camera on macosX, I'd be very grateful if you let me know.
The repository includes a submodule with the data, so to retrieve everything, you have to run
git clone --recursive https://github.com/idiap/hesm_distrib.git
or if you have cloned the repository without the submodules, from within the repo, run
git submodule init git submodule update
Physical setup - wiring
You need to compile the camera driver for
Python, to that end, run the following commands from within the
sudo apt-get install ueye python setup.py build_ext --inplace
This will install
ueye and compile a
Python wrapper for the
C++ code to drive the Thorlabs camera.
The wrapper only works for
Python2.7 and hasn't been translated to
Python3 (see why).
Note for Windows users : instead of
The code needs Jupyter notebook to be run. See here for installation instructions.
Because of the data acquisition code, the processing code is in
Python2.7 but should work with no modification with
Here is how to install the required python packages.
pip install numpy matplotlib colour_demosaicing tifffile pyqtgraph
pyqtgraph can be tricky/tenuous to install, if you can't or just don't want, it is easy to work without it; you will just have to select regions of interest "by hand".
We have a patent on the method presented in the paper, under the European Patent application number EP19154253.
Henceforth, the files distributed here are for non-commercial use only, see the license file within this repository. If you have a commercial interest in the present code, please contact our technology transfer office.