Example model for superresolution in EM

[esgomezm]

The superres_EM folder contains:

• config.yaml: configuration yaml file defined as similar as possible to the Bioimage Model Zoo
• exampleImage.tiff: input image for the model
• resultImage.tiff: ground truth for the exampleImage (what the model should output)
• saved_model.pb + variables: tensorflow trained model (2D U-Net)
• postprocessing.txt and preprocessing.txt: fiji macros for the pre/post processing (in this case, only a preprocessing is required)
• U_Net_4_superresolution_in_EM.ipynb: jupyter notebook / google colaboratory file with all the steps followed to train the model.

The material here was provided during the TS15 of the NEUBIAS training school, Bordeaux 2020, by Ignacio Arganda-Carreras.