Performs OCR on a list of images using Tesseract and performs fuzzy string matching with a given list of strings.
Tested against Windows 10 / Python 3.11 / Anaconda
pip install tesseractrapidfuzz
This function takes a path to the Tesseract OCR executable, a list of image paths or URLs,
a list of strings to compare against the recognized text, and optional fuzzy matching settings.
It returns a pandas DataFrame with OCR results and fuzzy matching scores.
Args:
tesseract_path (str): Path to the Tesseract OCR executable.
allpics (Union[list, tuple]): List of image paths, URLs, or other image data sources.
strings_to_compare (Union[list, tuple, np.ndarray]): List of strings for fuzzy matching.
compare_single_words (bool, optional): Enable fuzzy matching on individual words.
Defaults to True.
compare_grouped_words (bool, optional): Enable fuzzy matching on grouped words.
Defaults to True.
scorer_single_words (valid_scorer, optional): Fuzzy matching scorer for single words.
Defaults to "WRatio".
scorer_grouped_words (valid_scorer, optional): Fuzzy matching scorer for grouped words.
Defaults to "WRatio".
add_after_tesseract_path (str, optional): Additional arguments for Tesseract after
the input image path. Defaults to an empty string.
add_at_the_end (str, optional): Additional arguments to append to the Tesseract command.
Defaults to "-l eng --psm 3".
**kwargs: Additional keyword arguments to control the fuzzy matching process.
Returns:
pd.DataFrame: A DataFrame with OCR results and fuzzy matching scores, including columns:
- 'id_img': Image ID
- 'id_word': Word ID within the image
- 'ocr_result': Recognized text
- 'start_x': Starting X-coordinate of the bounding box
- 'end_x': Ending X-coordinate of the bounding box
- 'start_y': Starting Y-coordinate of the bounding box
- 'end_y': Ending Y-coordinate of the bounding box
- 'conf': Confidence score
- 'grouped_text': Grouped text for fuzzy matching
- 'compared_grouped_words_similarity': Fuzzy matching score for grouped words
- 'compared_grouped_words_index': Index of the matched string for grouped words
- 'compared_grouped_words_value': Matched value for grouped words
- 'compared_single_words_similarity': Fuzzy matching score for single words
- 'compared_single_words_index': Index of the matched string for single words
- 'compared_single_words_value': Matched value for single words
Example:
import re
from tesseractrapidfuzz import ocr_and_fuzzy_check
df = ocr_and_fuzzy_check(
tesseract_path=r"C:\Program Files\Tesseract-OCR\tesseract.exe",
allpics=[
"https://m.media-amazon.com/images/I/711y6oE2JrL._SL1500_.jpg",
"https://m.media-amazon.com/images/I/61g+KBpG20L._SL1500_.jpg",
],
strings_to_compare=[
"nonviolent",
"communication",
"emotional",
"well-being",
"terrible",
"today.",
"discover",
"definitive",
"guides",
"transforming",
"converting",
"conflict",
"meaningful",
"connection,",
"unveiling",
"inspirational",
"strategies",
"engagement.",
"martha",
"williams",
"nonviolent communication",
"emotional well-being",
"I had a terrible day at work today.",
"wait till you",
"heared about",
"the art of nonviolent communication",
"martha a. williams",
],
compare_single_words=True,
compare_grouped_words=True,
scorer_single_words="QRatio",
scorer_grouped_words="WRatio",
add_after_tesseract_path="",
add_at_the_end="-l eng --psm 3",
workers=5,
processor=lambda x: re.sub(r"\W+", "", str(x).lower()),
)
print(df.to_string())
# ...
# 7 1 8 terrible 448 563 371 396 77 2875 505 383 115 25 2 | had a terrible 100.000000 4 terrible 90.0 4 terrible
# 8 1 9 day 363 418 415 448 96 1815 390 431 55 33 3 day at work 75.000000 5 today. 90.0 22 I had a terrible day at work today.
# 9 1 10 at 427 457 418 440 96 660 442 429 30 22 3 day at work 50.000000 18 martha 90.0 22 I had a terrible day at work today.
# 10 1 11 work 466 540 415 440 96 1850 503 427 74 25 3 day at work 33.333332 6 discover 90.0 22 I had a terrible day at work today.
# 11 1 12 today. 402 498 460 492 96 3072 450 476 96 32 4 today. 100.000000 5 today. 100.0 5 today.
# 12 1 13 Wait 551 635 525 556 95 2604 593 540 84 31 5 Wait till you 53.333332 23 wait till you 100.0 23 wait till you
# 13 1 14 till 645 695 525 556 96 1550 670 540 50 31 5 Wait till you 53.333332 23 wait till you 100.0 23 wait till you
# 14 1 15 you 705 773 533 565 96 2176 739 549 68 32 5 Wait till you 42.857143 23 wait till you 100.0 23 wait till you
# 15 1 16 hear 562 645 579 610 95 2573 603 594 83 31 6 hear about 53.333332 24 heared about 90.0 24 heared about
# 16 1 17 about 663 767 579 610 96 3224 715 594 104 31 6 hear about 62.500000 24 heared about 90.0 24 heared about
# 17 2 1 ART 94 246 125 207 95 12464 170 166 152 82 7 ART OF NONVIOLENT 66.666664 18 martha 90.0 0 nonviolent
# 18 2 2 OF 275 376 125 207 95 8282 325 166 101 82 7 ART OF NONVIOLENT 40.000000 11 conflict 90.0 0 nonviolent
# 19 2 3 NONVIOLENT 407 907 125 206 96 40500 657 165 500 81 7 ART OF NONVIOLENT 100.000000 0 nonviolent 90.0 0 nonviolent
# 20 2 4 COMMUNICATION 167 832 296 377 96 53865 499 336 665 81 8 COMMUNICATION 100.000000 1 communication 100.0 1 communication
# 21 2 5 TAR 319 379 428 444 31 960 349 436 60 16 9 TAR 50.000000 5 today. 72.0 9 transforming
# 22 2 6 DISCOVER 192 307 624 667 96 4945 249 645 115 43 10 DISCOVER THE DEFINITIVE GUIDES TO NONVIOLENT 100.000000 6 discover 90.0 0 nonviolent
# 23 2 7 THE 320 360 624 667 96 1720 340 645 40 43 10 DISCOVER THE DEFINITIVE GUIDES TO NONVIOLENT 44.444443 18 martha 90.0 0 nonviolent
# 24 2 8 DEFINITIVE 374 507 624 667 96 5719 440 645 133 43 10 DISCOVER THE DEFINITIVE GUIDES TO NONVIOLENT 100.000000 7 definitive 90.0 0 nonviolent
# 25 2 9 GUIDES 521 604 624 667 96 3569 562 645 83 43 10 DISCOVER THE DEFINITIVE GUIDES TO NONVIOLENT 100.000000 8 guides 90.0 0 nonviolent
# 26 2 10 TO 618 645 628 654 96 702 631 641 27 26 10 DISCOVER THE DEFINITIVE GUIDES TO NONVIOLENT 57.142857 5 today. 90.0 0 nonviolent
# 27 2 11 NONVIOLENT 661 810 624 667 96 6407 735 645 149 43 10 DISCOVER THE DEFINITIVE GUIDES TO NONVIOLENT 100.000000 0 nonviolent 90.0 0 nonviolent
# ...
Note:
- The function combines OCR results with fuzzy string matching, allowing for versatile text analysis.
- Valid_scoring options are: "WRatio", "QRatio", "ratio", "partial_ratio".
1 maintainer
Metadata
1 maintainer
Release files for tesseractrapidfuzz 0.10
For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.
Source distribution (sdist)
| File | Size | Uploaded | Provenance | Actions |
|---|---|---|---|---|
| tesseractrapidfuzz-0.10.tar.gz | 26.9 kB | Without provenance | Details for tesseractrapidfuzz-0.10.tar.gz |
Built distribution (wheel)
| File | Interpreter | ABI | Platform | Actions Reset |
|---|---|---|---|---|
| tesseractrapidfuzz-0.10-py3-none-any.whl | Python 3 | none | any | Details for tesseractrapidfuzz-0.10-py3-none-any.whl |
Total release size: 55.7 kB
Release files / tesseractrapidfuzz-0.10.tar.gz
| Download URL | tesseractrapidfuzz-0.10.tar.gz |
|---|---|
| Size | 26.9 kB |
| Tags | Source |
|
SHA-256 checksum How to use checksums |
494a5494477205b3a4f0c4b195fc34f182ab4d99a2ea74181cb276bb43df96dc
|
|
BLAKE2b-256 checksum How to use checksums |
184f4a726a3d943e608351a7d3e9cbfdf6a154eaf81d5e9ba2ea5316c5415c74
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.5
|
Release files / tesseractrapidfuzz-0.10-py3-none-any.whl
| Download URL | tesseractrapidfuzz-0.10-py3-none-any.whl |
|---|---|
| Size | 28.8 kB |
| Tags | Python 3 |
|
SHA-256 checksum How to use checksums |
50201d9f7f9837dd060552c2e60cb2d593422900e2ebcc7b30148d1fbedc72e9
|
|
BLAKE2b-256 checksum How to use checksums |
32c92156ab989edbbd8a2ccc54032afb6826827ba27412b36d26f344a1e7cac2
|
| Upload date | |
|
Uploaded using Trusted Publishing? What is trusted publishing? |
No |
| Uploaded via |
twine/4.0.2 CPython/3.11.5
|
Security for tesseractrapidfuzz 0.10
Provenance
Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.
No provenance found
None of the files in this release were uploaded with provenance metadata.
A note on package security
PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.
To learn more about protecting yourself from malicious actors, visit our package security documentation.
