NNExtractor scores all images in the folder

WFP-VAM / HRM

High Resolution Mapping of Food Security

https://wfp-vam.github.io/HRM/

MIT License

21 stars 6 forks source link

Closed lorenzori closed 6 years ago

lorenzori commented 6 years ago

we should make it feasible to score only a subset of images, for example in a production environment to score only relevant pictures.

lorenzori commented 6 years ago

something like we did in the _average_feature_dir:

i = int(name[0: 5]) + a
j = int(name[6:10]) + b
img_path = os.path.join(image_dir, name, str(i)+'_'+str(j)+".jpg")

lorenzori commented 6 years ago

we could always put the logic outside, i.e. specify a different directory or loop over directories of images to process...

pasquierjb commented 6 years ago

extract_features now takes list_i,list_j as input