ocrmypdf-paddleocr 0.1.1

PaddleOCR engine plugin for OCRmyPDF

OCRmyPDF PaddleOCR

A plugin to use PaddleOCR as the OCR engine for OCRmyPDF, instead of Tesseract. PaddleOCR generally provides better accuracy, especially for CJK languages (Chinese, Japanese, Korean).

Unlike the EasyOCR plugin, this plugin does not require Tesseract for any operations.

Installation

pip install ocrmypdf-paddleocr

Usage

ocrmypdf --plugin ocrmypdf_paddleocr input.pdf output.pdf
ocrmypdf --plugin ocrmypdf_paddleocr -l kor input.pdf output.pdf

Or from Python:

import ocrmypdf

ocrmypdf.ocr('input.pdf', 'output.pdf', plugins=['ocrmypdf_paddleocr'])

Known limitations

• The plugin forces jobs=1. PaddlePaddle's inference engine already uses all CPU cores internally, so parallel page processing would only cause contention

License

MPL-2.0