Brumm, Pauline
Pauline
Brumm
0000-0002-8220-0676
Ciotta, Nicola
Nicola
Ciotta
0000-0001-6291-1967
Sauer, Hans Martin
Hans Martin
Sauer
Blaeser, Andreas
Andreas
Blaeser
Doersam, Edgar
Edgar
Doersam
0000-0002-4338-1777
TU Darmstadt
05n911h24
Python code: Classification of in situ high speed videos of the gravure printing fluid splitting process using deep learning
TU Darmstadt
2022
gravure printing
fluid splitting
Python code
deep learning
video classification
pattern formation
machine learning
viscous fingering
403-03 Mechanische Verfahrenstechnik
403-03 Mechanical Process Engineering
404-03 Strömungsmechanik
404-03 Fluid Mechanics
405-03 Beschichtungs- und Oberflächentechnik
405-03 Coating and Surface Technology
660
620
TU Darmstadt
2022-08-26
2022-08-26
2022
en
Text
https://tudatalib.ulb.tu-darmstadt.de/handle/tudatalib/3552
https://doi.org/10.48328/tudatalib-938
10.1007/s11998-022-00687-x
10.25534/tudatalib-191.2
978-1-83864-483-3
https://github.com/PacktPublishing/PyTorch-Computer-Vision-Cookbook
https://github.com/ladisk/pyMRAW
Creative Commons Attribution-NonCommercial 4.0
The files that are made available here are all the key components that are needed to recreate the automated classification of high speed videos of the gravure printing fluid splitting process using deep learning. The purpose of this was to create a tool that can distinguish the distinct regimes of pattern formation. This could ultimately lead to a better understanding of the fluid splitting process and a complete map of pattern formation regimes. <br />
<br />
The zip-file "Python_code" contains the Python script that was used to extract PNG frames from a video dataset created by Julian Schäfer in 2019 (<a href="https://doi.org/10.25534/tudatalib-191.2">DOI: 10.25534/tudatalib-191.2</a>). The result of this frame extraction is provided as "Frame_dataset_PNG_8_bit.zip". Additionally, "Python_code.zip" contains the scripts for training the deep learning models and using them for inference. All trained models for the associated publication can be found in the zip-file "models".<br />
<br />
Further information can be found in the README-file and in the publication:<br />
Pauline Brumm, Nicola Ciotta, Hans Martin Sauer, Andreas Blaeser, Edgar Dörsam, 2022. Deep learning study of induced stochastic pattern formation in the gravure printing fluid splitting process, Journal of Coatings Technology and Research. <a href="https://doi.org/10.1007/s11998-022-00687-x">DOI: 10.1007/s11998-022-00687-x</a>
DFG
SFB1194
TP C01 Dörsam