A convolutional neural network architacture for permeability upscaling
I designed and coded this network for a research paper.
The code is written in Python and requires the following packages: tensorflow, scipy, numpy and matplotlib.
The code is a single file.
In addition to the code, an example, created based on Strebelle’s 250×250 fluvial channel training image is included, is included. For further information, please see the following paper.
@article{eage:/content/papers/10.3997/2214-4609.202244018, author = "Sayyafzadeh, M. and Guérillot, D.", title = "Rapid Permeability Upscaling using Convolutional Neural Networks", journal= "", year = "2022", volume = "2022", number = "1", pages = "1-12", doi = "https://doi.org/10.3997/2214-4609.202244018", url = "https://www.earthdoc.org/content/papers/10.3997/2214-4609.202244018", publisher = "European Association of Geoscientists & Engineers", issn = "2214-4609", type = "", }
Please cite the above paper if you use this code for your work.