Tesseract

lastmod 2025-02-02created 2017-06-2115 edits

Tesseract Open Source OCR Engine

Categories

• pytesseract (Python binding)
• OCR
• LLM-Aided OCR - Fixing Tesseract OCR errors with an LLM

How to install

$ sudo apt-get install tesseract-ocr

## for dev
$ sudo apt-get install libtesseract-dev

Pre-trained models

Place the trained data in the /usr/share/tesseract-ocr/4.00/tessdata directory.

• Github - tesseract-ocr/tessdata - Trained models with support for legacy and LSTM OCR engine
• Models for OCR-D processors
• Training Tesseract 4 models from real images | End Point
• Github - Shreeshrii/tessdata_shreetest - finetuned traineddata files for tesseract 4.0.0 for testing
  • digits1.traineddata - OCR model applied for LS ELECTRIC
  • Use the digits.traineddata file.
• Github - tesseract-ocr/tessdata_best - Best (most accurate) trained LSTM models.

See also

• OCR
• OpenALPR
• Leptonica

Favorite site

• Github - Tesseract project site
• Python - Machine learning project (1) - Extracting text from OCR images (Tesseract) :: A Good Person's Dev Notes
• OCR Korean training

Article

• Our Search for the Best OCR Tool, and What We Found