datadings.sets.Vaihingen_write module

usage: Vaihingen_write.py [-h] [-o PATH] [-y] INDIR

Create Vaihingen data set files.

This tool will look for the ISPRS_semantic_labeling_Vaihingen directory in the input directory.

See also

http://www2.isprs.org/commissions/comm3/wg4/2d-sem-label-vaihingen.html

Note

Requests to download the dataset can be made by filling out the form on the website.

Warning

Images, labels, and masks are numpy arrays, not images!

Important

Samples have the following keys:

  • "key"

  • "image"

  • "label_image"

  • "mask"

Positional arguments

INDIR Directory that contains dataset source files.

options:
-h, --help

show this help message and exit

-o PATH, --outdir PATH

Output directory. Defaults to indir.

-y, --no-confirm

Don’t require user interaction.

datadings.sets.Vaihingen_write.images_label_dsm_iter(indir, image_ids)[source]
datadings.sets.Vaihingen_write.main()[source]
datadings.sets.Vaihingen_write.map_color_values_to_class_indices(img)[source]
datadings.sets.Vaihingen_write.write(writer, img, labels, mask, filename)[source]
datadings.sets.Vaihingen_write.write_sets(indir, outdir, args, crop_size=(384, 384))[source]