# YCG09/chinese_ocr

CTPN + DenseNet + CTC based end-to-end Chinese OCR implemented using tensorflow and keras

Repository: https://github.com/YCG09/chinese_ocr
Canonical: https://ross.abutalabs.com/products/chinese_ocr
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2019-10-08T03:48:03+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": 3092, "days_push": 2521, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2782, forks 1065 (observed 2026-08-28T04:07:21.746456+00:00)

## What it is
An end-to-end Chinese OCR system implemented with TensorFlow and Keras, combining CTPN for text detection with DenseNet + CTC for text recognition. It supports variable-length Chinese character detection and recognition, and includes training scripts and a synthetic dataset of ~3.64 million images.

## Use cases
- recognize chinese text in images
- extract text from photos of chinese documents
- detect and read chinese characters from screenshots
- build an ocr pipeline for chinese pdfs or scanned pages
- train a custom chinese text recognition model
- run offline ocr on chinese signage images

## When to choose
- you need open-source Chinese OCR with trainable models
- you want a TensorFlow/Keras-based OCR you can retrain on your own data
- you need both text detection and recognition in one pipeline

## When to avoid
- you need actively maintained software with recent updates
- you need OCR for many languages beyond Chinese and English
- you prefer a production-ready OCR service like PaddleOCR or Tesseract with broader support

## Facets
- artifact type: library
- maturity: maintenance
- function: ocr, computer-vision, image-processing, machine-learning, deep-learning
- domain: computer-vision, image-processing, machine-learning
- platform: python, windows
- tags: chinese-ocr, ctpn, densenet, ctc, text-detection, text-recognition, tensorflow, keras, natural-language-processing, linux, macos

## Member repositories
- YCG09/chinese_ocr (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:21.746456+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 2026-08-30T08:16:04.688560+00:00, confidence not recorded.
  - readme: https://github.com/YCG09/chinese_ocr (fetched 2026-08-28T04:07:21.746456+00:00, sha 0ddff7b98e1b)
- Data as of 2026-08-30T08:39:29.467469+00:00.