# AstarLight/CPS-OCR-Engine

An awesome OCR engine developed by SYSU DeepDriving Lab

Repository: https://github.com/AstarLight/CPS-OCR-Engine
Canonical: https://ross.abutalabs.com/products/cps-ocr-engine
Language: Python
License Family: other
Last push: 2019-05-11T08:12:39+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3193, "days_push": 2671, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1083, forks 501 (observed 2026-08-28T04:03:31.140842+00:00)

## What it is
A deep-learning-based OCR engine from SYSU DeepDriving Lab that recognizes 3755 printed Chinese characters (Level-1 character set) in electronic documents, IDs, and tickets. It includes synthetic data generation, training, evaluation, and inference scripts in Python.

## Use cases
- recognize printed chinese characters in scanned documents
- build an ocr pipeline for invoices and receipts
- train a custom chinese character recognition model
- extract text from id cards and tickets
- generate synthetic chinese font datasets for training
- replace paid ocr apis with a self-hosted model

## When to choose
- you need offline, self-hosted OCR for printed Chinese text
- you want to fine-tune or retrain the recognition model yourself
- you're processing Chinese documents, tickets, or certificates

## When to avoid
- you need handwriting or general multilingual OCR
- you need a maintained project with a license and active releases
- you need production-grade support or line-level text segmentation out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: ocr, machine-learning, deep-learning, image-processing
- domain: computer-vision, machine-learning
- platform: python, cross-platform
- tags: chinese-ocr, printed-text-recognition, tensorflow, invoice-recognition, document-processing, no-license, natural-language-processing, linux

## Member repositories
- AstarLight/CPS-OCR-Engine (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:31.140842+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 2026-08-30T06:50:11.950261+00:00, confidence not recorded.
  - readme: https://github.com/AstarLight/CPS-OCR-Engine (fetched 2026-08-28T04:03:31.140842+00:00, sha a1d9ced9f3ee)
- Data as of 2026-08-30T08:39:29.467469+00:00.