Wrapper to pytesseract to preserve space and formatting
Project description
OCR_with_format
- simple wrapper that postprocesses pytesseract's hOCR output to maintain format and spacings.
- Link to an alternative implementation found on stackoverflow
How to
- install
python -m pip install OCR_with_format
- see usage
OCR_with_format --help
(executing withpython -m
is not supported)
Usage
NAME
OCR_with_format
SYNOPSIS
OCR_with_format IMG_PATH <flags>
POSITIONAL ARGUMENTS
IMG_PATH
Type: str
path to the image you want to do OCR on
FLAGS
-m, --method=METHOD
Type: str
Default: 'with_format'
if 'with_format', will use the author's code if 'none', will output tesseract's default output if 'stackoverflow', will output using another algorithm found on stackoverflow
--thresholding_method=THRESHOLDING_METHOD
Type: str
Default: 'otsu'
any from "otsu", "otsu_gaussian", "adaptative_gaussian", "all"
If "all", the three methods will be tried and the final output will be the one which maximizes the mean and median confidences over each parsed words.
-l, --language=LANGUAGE
Type: str
Default: 'eng'
language to look for in the image
-o, --output_path=OUTPUT_PATH
Type: Optional[str]
Default: None
if not None, will output to this path and erase its previous content.
--tesseract_args=TESSERACT_ARGS
Type: str
Default: '-...
default arguments for tesseract
-q, --quiet=QUIET
Default: False
if True, will only print the output and no logs
NOTES
You can also use flags syntax for POSITIONAL ARGUMENTS
Example
- Image:
- output from
OCR_with_format ./screenshot.png --thresholding_method="all" --quiet
@Unwateh (1) ~ Fork (3) (©
OCR_with_format [ Pir ][ | [ & <) [ s -]
About
&
Wrapper around pytesseract to
¥ Branches © Tags postprocess in a way that preserves
spacing and formattings.
i thiswillbeyourgithub addded license 10 minutes ago O 4
&5 GPL-3.0 license
@ LICENSE addded license 10 minutes ago - Activity
[u] __init__.py minor 11 minutes ago ¢ Ostars
[u] requirements.txt added empty requirements 11 minutes ago <& 1 watching
Y Oforks
Help people interested in this repository understand your project by
adding a README. Releases
Create No releases a new published release
Packages
No packages published
Publish your first package
Languages
———
® Python 100.0%
- output from
OCR_with_format ./screenshot.png --quiet --comparison_run
*
OCR_with_format
[ pin | [ @unwateh (@) ~ | [ & Fork (O)
-] [ ¢ s (0
¥ main ~
¥ Branches © Tags
Wrapper around pytesseract to
postprocess in a way that preserves
. . . spacing and formattings.
- thiswillbeyourgithub addded license 10 minutes ago 'O 4
&5 GPL-3.0 license
@ LICENSE addded license 10 minutes ago A Activity
0O _init__py minor 11 minutes ago ¢ Ostars
O requirements.txt added empty requirements 11 minutes ago | @ 1watching
% 0forks
Help people interested in this repository understand your project by
adding a README. Releases
No releases published
Create a new release
Packages
No packages published
Publish your first package
Languages
————
@ Python 100.0%
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
ocr_with_format-0.12.tar.gz
(19.8 kB
view hashes)
Built Distribution
Close
Hashes for OCR_with_format-0.12-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | b968663df8c4bae8b25c49c773ed8b559d1be29f000ac6af7c5a99700d233df9 |
|
MD5 | e785f92df98def4cade76f8bd37bb8fb |
|
BLAKE2b-256 | 28779f14a4a005993b44e0bab6a67d1658162210ee94fdd934e56245054c31c5 |