Skip to main content

Text Recognition OCR/HTR (Handwritten Text Recognition)

dhSegment  segmentation and layout analysis for historical documents. Documentation: Read the Docs

docTR  OCR with detection and recognition, designed for easy project integration. Documentation: mindee.github.io/doctr

eScriptorium  an open-source web platform for managing transcription workflows (import, layout segmentation, transcription, correction, and model training), built on top of Kraken. Documentation: Read the Docs

Kraken  an OCR/HTR engine with layout analysis. Documentation: kraken.re

LayoutParser  layout detection (blocks/regions) and utilities for building pipelines. Documentation: Read the Docs

PaddleOCR  OCR and document parsing (structure and layout). Documentation: paddleocr.ai

Tesseract OCR  an open-source OCR tool that is especially strong for printed text. Documentation: tessdoc

Transkribus  platform for printed text recognition (OCR) and handwritten text recognition (HTR) in historical documents.