PyPI Stats

Search

All packages
Top packages

Track packages

abstract-ocr


PyPI page
Home page
Author: putkoff
Summary: A structured OCR pipeline designed for layout-aware text extraction from complex documents, combining preprocessing, column detection, region classification, PaddleOCR, and ordered OCR assembly.
Latest version: 0.0.1.70
Required dependencies: abstract_pdfs | abstract_utilities | beautifulsoup4 | lxml | moviepy | nltk | numpy | opencv-python-headless | openpyxl | packaging | paddleocr | paddlepaddle | paddlex | pdf2image | pdfminer.six | pdfplumber | pillow | pymupdf | pypdf2 | pypdfium2 | python-docx | pyyaml | requests | scikit-image | spacy | tqdm | typing_extensions
Optional dependencies: abstract_hugpy | easyocr | pytesseract

Downloads last day: 218
Downloads last week: 3,085
Downloads last month: 4,727