Metadata-Version: 2.1
Name: lineaug
Version: 0.1
Summary: Augment line images for improving OCR datasets 
Home-page: https://github.com/maxnth/lineaug
Author: Maximilian Nöth
Author-email: maximilian.noeth@protonmail.com
License: MIT License
Description: # LineAug
        Augment line images for improving OCR datasets 
        
        ## CLI
        ```
        usage: augment.py [-h] -i [IMAGES [IMAGES ...]] [-o OUTPUT] [-gt GROUND_TRUTH] [-n N] [-bg] [-e]
        
        Augment OCR in the form of line images.
        
        optional arguments:
          -h, --help            show this help message and exit
          -i [IMAGES [IMAGES ...]], --images [IMAGES [IMAGES ...]]
                                Path to line image(s).
          -o OUTPUT, --output OUTPUT
                                Output path where augmented images will be saved.
          -gt GROUND_TRUTH, --ground_truth GROUND_TRUTH
                                Extension of the ground truth text files. Will create new ground truth files for the augmented line images containing the existing ground truth for the associated
                                line (optional).
          -n N                  Number of augmented line image variants to create for each input.
          -bg, --background     Whether to add noise to the background of the line image.
          -e, --enumerate       Enumerate output file names instead of using input file names.
        ```
        
Keywords: Augmentation,OCR,optical character recognition
Platform: UNKNOWN
Classifier: Programming Language :: Python :: 3
Classifier: License :: OSI Approved :: MIT License
Classifier: Operating System :: OS Independent
Requires-Python: >=3.6
Description-Content-Type: text/markdown
