PreprocessTrainingData.m

This script takes a stack of images and labels performs necessary conversions so they can be used as training data for CDeep3M.

Usage:

Use -> PreprocessTrainingData /ImageData/training/images/ /ImageData/training/labels/ /ImageData/augmentedtraining/

Images and labels should be in format described here

Description of format of .h5 output files can be found here

Contents of output directory generated by PreprocessTrainingData.m:

augmentedtraining/
├── training_full_stacks_v10.h5
├── training_full_stacks_v11.h5
├── training_full_stacks_v12.h5
├── training_full_stacks_v13.h5
├── training_full_stacks_v14.h5
├── training_full_stacks_v15.h5
├── training_full_stacks_v16.h5
├── training_full_stacks_v1.h5
├── training_full_stacks_v2.h5
├── training_full_stacks_v3.h5
├── training_full_stacks_v4.h5
├── training_full_stacks_v5.h5
├── training_full_stacks_v6.h5
├── training_full_stacks_v7.h5
├── training_full_stacks_v8.h5
└── training_full_stacks_v9.h5