Add processed data

Each data is in the tuple form of (image_name, class)
In each folder, there are 
1)  all processed mask of a category. e.g. Binary_map_aug/train/1/
2) all image names that contain this category. e.g. Binary_map_aug/train/1.txt 
3) all data for training and validation. e.g.  e.g. Binary_map_aug/train/split0_train.txt