Train VGG16 for binary image-classification

Train a deep-learning model for image-classification into 2 target categories.
The training is done by transfer learning of a VGG-16 base pretrained on the ImageNet dataset, completed by freshly initialized fully connected classification layers.

Category ground-truth annotations should be done in Fiji using the Qualitative-Annotation plugins, see reference below.

If you use this workflow please cite :

Thomas LSV, Schaefer F and Gehrig J.
Fiji plugins for qualitative image annotations: routine analysis and application to image classification
[version 1; peer review: awaiting peer review].
F1000Research 2020, 9:1248
doi: 10.12688/f1000research.26872.1

External resources

• Publication
• YouTube tutorial
• Original workflow
• Example dataset
• Online Documentation (GitHub)