ACsN (pronounced as action) stands for Automatic Correction of sCMOS-related Noise. It combines an accurate estimation of noise variation with sparse filtering to eliminate the most relevant noise sources in the images of a sCMOS sensor. This results in a drastic reduction of pixel-dependent noise in sCMOS images and an enhanced stability of denoising performance at a competitive computational speed.

Please, cite our paper on Nature Communications.

Mandracchia, B., Hua, X., Guo, C. et al. Fast and accurate sCMOS noise correction for fluorescence microscopy. Nat Commun 11, 94 (2020) doi:10.1038/s41467-019-13841-8

• Extended support to Linux and Mac OS (partial)
• Added Python version
• Addition of weight factor to allow for user control of smoothing
• Video filtering processes have been updated
• Sparse Filtering updated
• Python version is partially compatible with GPU computing

ACsN requires a standard computer with enough RAM to support Python >= 3.7. For minimum performance, this will be a computer with about 2 GB of RAM. For optimal performance, we recomend the following specs:

RAM: 16+ GB; CPU: 6+ cores, 3.2+ GHz/core.

Python 3.7+ Windows OS 7+ Linux OS Partial functionality on Mac OS

To run ACsN files:

• Clone this repository
• Run the command 'python setup.py install' after you're in the Sparse_Filtering folder
• Install VapourSynth from https://github.com/vapoursynth/vapoursynth/releases
  • Install the R48 version if using Python 3.7. Otherwise, install the newest version
  • Once installed, got to the directory where vsrepo.py is located and install bm3d and msvfunc using the commands:
    • vsrepo.py install bm3d
    • vsrepo.py install msvfunc
• Load your files using the ASCN_Run.py file. Run the ACSN_Run.py file in the terminal using the command (possible only when you're in the same directory):
  • python ACSN_Run.py

ACsN requires a standard computer with enough RAM to support MATLAB 2018b. For minimum performance, this will be a computer with about 4 GB of RAM. For optimal performance, we recomend the following specs:

RAM: 16+ GB; CPU: 6+ cores, 3.2+ GHz/core.

MATLAB 2018b+

MATLAB "Curve Fitting" Toolbox

Windows OS 64 bit, Linux 64 bit or Mac OS X 64 bit*

To run ACsN from MATLAB command line:

• Add the folder ACsN_code to your MATLAB path (including subfolders).
• In the command line type help ACSN or run the Sample code script in the Test Images folder to see the code usage.

To run ACsN from ImageJ/Fiji follow these steps:

• Add the ImageJ-MATLAB update site to ImageJ. To see how, look at here.
• Go to Edit > Options > MATLAB and enter the file path for MATLAB licence.
• Add the ACsN_code folder and subfolders to the MATLAB path.
• Copy the file 'ACsN_.m' to the folder '\plugins\Scripts\Process'.
• Select an open image in ImageJ and then press Process > ACsN from the menu toolbar.
• To test the program you can use the images provided in the Test Images folder. See the file Settings.txt for the aquisition parameters.

The installation on a recommended computer should take less than 3 seconds.

* Mac OS is only partially supported

Suraj Rajendran (Python Version) and Biagio Mandracchia