tiramisu_keras

This repository is the implementation (keras) of:

Xiao M., Rothermel M., Tom M., Galliani S., Baltsavias E., Schindler K.: Lake Ice Monitoring with Webcams, ISPRS Annals of the Photogrammetry, Remote Sensing and Spatial Information Sciences, IV-​2, pages 311-​317, 2018

This work is part of the Lake Ice Project (Phase 1). Here is the link to Phase 2 of the same project.

The implementation was extended and modified starting from the one in 0bserver07/One-Hundred-Layers-Tiramisu .

Pre-requisites:

• Numpy
• Keras
• H5py
• Opencv
• Tensorflow

Directory structure:

+ tiramisu_keras
+ Data
 + Images
 + Labels
+ Model
+ Result

Data preparation

1. Write the image data files in to an HDF5 file: python saveHDF5.py PATH_TO_IMAGE PATH_TO_ANNOTATION PATH_TO_HDF5 NAME_TO_HDF5 (e.g. python saveHDF5.py ../Data/Images/ ../Data/Labels/ ../Data/ demo.hdf5)

2. Divide the image data into training, validation and testing data sets: python data_loader.py PATH_TO_IMAGE --hdf5_dir PATH_TO_HDF5 --hdf5_file NAME_TO_HDF5 (e.g. python data_loader.py ../Data/Images/ --hdf5_dir ../Data/ --hdf5_file demo.hdf5)

Training and evluation

python train.py PATH_TO_IMAGE PATH_TO_HDF5 NAME_OF_HDF5_FILE --dim_patch PATCH_SIZE --pre_trained PATH_NAME_PRETRAINED_MODEL

Testing

python test.py PATH_TO_STORE_PREDICTION NAME_OF_THE_MODEL DATA_FILE_NAME --dim_patch PATCH_SIZE

Citation

@Article{muyan_lakeice_2018, AUTHOR = {Xiao, M. and Rothermel, M. and Tom, M. and Galliani, S. and Baltsavias, E. and Schindler, K.}, TITLE = {Lake ice monitoring with webcams}, JOURNAL = {ISPRS Annals of Photogrammetry, Remote Sensing and Spatial Information Sciences}, VOLUME = {IV-2}, YEAR = {2018}, PAGES = {311--317}, DOI = {10.5194/isprs-annals-IV-2-311-2018} }