# Calamari-OCR/calamari

Line based ATR Engine based on OCRopy

Repository: https://github.com/Calamari-OCR/calamari
Canonical: https://ross.abutalabs.com/products/calamari
Language: Python
License: GPL-3.0
License Family: copyleft
Last push: 2026-06-23T22:11:57+00:00

## Health v2 (maintenance only)
Score: 74/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 89, release rhythm 40, longevity 100
- inputs: {"age_days": 3088, "days_push": 71, "days_rel": 659, "gap_med": 20.5, "n_releases_24m": 3}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1197, forks 210 (observed 2026-08-28T04:03:57.394284+00:00)

## What it is
Calamari is a Python-based OCR engine for line-based automatic text recognition, built on OCRopy and Kraken with a TensorFlow deep-learning backend. It offers both a command-line interface and a modular Python API, with pretrained models available for many historical scripts.

## Use cases
- recognize text in scanned historical documents
- train a custom OCR model for old prints
- run OCR on line images from the command line
- integrate OCR into a Python digitization pipeline
- recognize handwriting in manuscripts
- ensemble multiple OCR models for better accuracy

## When to choose
- you need trainable, high-accuracy OCR for historical or non-Latin scripts
- you want a scriptable Python OCR engine with pretrained models
- you work with line-segmented images from OCRopy/Kraken pipelines

## When to avoid
- you need general-purpose document layout analysis out of the box
- you want a GUI-based OCR tool
- you need OCR for modern documents where Tesseract suffices

## Facets
- artifact type: library
- maturity: active
- function: ocr, machine-learning, deep-learning, cli
- domain: computer-vision, image-processing, developer-tools
- platform: python, cli, cross-platform
- tags: ocr-engine, ocropy, kraken, tensorflow, handwritten-text-recognition, historical-documents, line-based-ocr, natural-language-processing

## Member repositories
- Calamari-OCR/calamari (main) score 74

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:57.394284+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:21:16.635570+00:00, confidence not recorded.
  - readme: https://github.com/Calamari-OCR/calamari (fetched 2026-08-28T04:03:57.394284+00:00, sha c9485e4f4c77)
- Data as of 2026-08-30T08:39:29.467469+00:00.