https://github.com/bionanoimaging/cellstorm-matlab
Code to generate datasets for the cellSTORM approach
Science Score: 23.0%
This score indicates how likely this project is to be science-related based on various indicators:
-
○CITATION.cff file
-
✓codemeta.json file
Found codemeta.json file -
○.zenodo.json file
-
○DOI references
-
✓Academic publication links
Links to: arxiv.org -
○Committers with academic emails
-
○Institutional organization owner
-
○JOSS paper metadata
-
○Scientific vocabulary similarity
Low similarity (11.4%) to scientific vocabulary
Repository
Code to generate datasets for the cellSTORM approach
Basic Info
- Host: GitHub
- Owner: bionanoimaging
- Language: Matlab
- Default Branch: master
- Size: 57 MB
Statistics
- Stars: 5
- Watchers: 5
- Forks: 1
- Open Issues: 0
- Releases: 0
Metadata Files
README.md
cellSTORM MATLAB Files
Based on paper by Benedict Diederich, Patrick Then, Alexander Jügler, Ronny Förster, Rainer Heintzmann
Introduction
The Matlab files are a collection of functions to prepare the training data as well as evaluation the results. It's a not-ordered list of files which were useful when training the GAN in Tensorflow. The major files will get a further explanation.
All files were tested on MATLAB 2016b.
Simulating training
To produce training data we follow two different approaches. We rely on purely simulated data coming from ThunderSTORM. Addtionally we introduce experimental data from cellSTORM measurements coming from the Huawei P9. We localize it with TunderSTORM and then form datapairs.
The datapairs are consisting groundtruth (GT) and degraded (i.e. measured, psf-convolved+noisy) frames. The GAN then tries to go back from degraded frames to localization maps (i.e. found center spot of a possible blinking fluorophore.).
For a very good tutorial on how to prepare the TunderSTORM dataset please also have a look at the DeepSTORM repository here.
Create Data from measurements
The file B_Generate_PSF_from_Measurement.m produces datapairs inside a .h5 file which can be read by the neural network. The software is a derivation of the work by Nehme et al. 2018. Our first attempts were much slower, but are still available in the /Functions folder.
For further processing, you need to place the creatd .h5 holding the datapairs (GT-degraded frames) into the GAN folder. Therfore please go to the follow-up repository here.
Benchmark Dataset
To Benchmark the Software, we created a set of files with varying compression rate as well as number of photons. These files can be found in the GoogleDrive
Example videos
To process data, please also download the example videos from the GoogleDrive.
DISCLAIMER
The Code will be provided as is whithout any guarantee of correct functionality. It's more a proof-of-principle rather than a working solution. If you find any errors, feel free to comment on them or fork the repository to solve the issues.
Owner
- Name: Nanoimaging at the IPHT Leibniz Institut Jena
- Login: bionanoimaging
- Kind: organization
- Location: Jena, Germany
- Website: nanoimaging.de
- Repositories: 33
- Profile: https://github.com/bionanoimaging
GitHub Events
Total
- Watch event: 2
Last Year
- Watch event: 2
Committers
Last synced: about 1 year ago
Top Committers
| Name | Commits | |
|---|---|---|
| beniroquai | b****d@g****e | 8 |
Committer Domains (Top 20 + Academic)
Issues and Pull Requests
Last synced: about 1 year ago
All Time
- Total issues: 0
- Total pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Total issue authors: 0
- Total pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0
Past Year
- Issues: 0
- Pull requests: 0
- Average time to close issues: N/A
- Average time to close pull requests: N/A
- Issue authors: 0
- Pull request authors: 0
- Average comments per issue: 0
- Average comments per pull request: 0
- Merged pull requests: 0
- Bot issues: 0
- Bot pull requests: 0