# xiaofengShi/CHINESE-OCR

End-to-end Chinese scene-text detection and recognition with CTPN, CRNN, and CTC (legacy project).

Repository: https://github.com/xiaofengShi/CHINESE-OCR
Canonical: https://ross.abutalabs.com/products/chinese-ocr
Homepage: https://xiaofengshi.com/2019/01/23/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0-TextDetection/
Language: Python
License Family: other
Topics: pytorch-crnn, keras-crnn, tensorflow, ocr, lstm-ctc, ctpn, chinese-ocr, crnn, legacy-project, scene-text-recognition
Last push: 2026-08-21T13:00:32+00:00

## Health v2 (maintenance only)
Score: 76/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 98, release rhythm 35, longevity 100
- inputs: {"age_days": 3063, "days_push": 12, "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 2955, forks 941 (observed 2026-08-28T04:07:32.163344+00:00)

## What it is
An end-to-end Chinese scene-text OCR pipeline combining CTPN for text detection, a VGG16-based orientation classifier, and CRNN with CTC for variable-length recognition, implemented in TensorFlow/Keras and PyTorch. It is a legacy research project kept for reproduction and reference, with no active feature development and no explicit open-source license.

## Use cases
- recognize Chinese text in natural scene images
- detect text regions in photos with CTPN
- identify text orientation 0/90/180/270 degrees
- train a CRNN CTC model for variable-length OCR
- reproduce classic scene-text detection and recognition research

## When to choose
- you need a reference implementation of CTPN + CRNN + CTC for Chinese OCR
- you are studying or reproducing 2016-2019 era scene-text recognition methods
- you need trainable Keras or PyTorch CRNN code for research

## When to avoid
- you need a maintained, production-ready OCR system
- you require a clear open-source license for commercial use
- you cannot pin old dependencies like Python 3.6 and TensorFlow 1.x
- you need vertical or rotated text detection out of the box

## Facets
- artifact type: library
- maturity: maintenance
- function: ocr, computer-vision, deep-learning, image-processing
- domain: computer-vision, image-processing, deep-learning
- platform: python
- tags: chinese-ocr, ctpn, crnn, ctc, scene-text-recognition, tensorflow, keras, pytorch, legacy-project, text-detection, natural-language-processing, linux, gpu

## Member repositories
- xiaofengShi/CHINESE-OCR (main) score 76

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:32.163344+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 2026-08-30T07:32:36.159162+00:00, confidence not recorded.
  - readme: https://github.com/xiaofengShi/CHINESE-OCR (fetched 2026-08-28T04:07:32.163344+00:00, sha cd2c744ac247)
  - homepage: https://xiaofengshi.com/2019/01/23/%E6%B7%B1%E5%BA%A6%E5%AD%A6%E4%B9%A0-TextDetection/ (fetched 2026-08-29T09:47:42.755740+00:00, sha 437376527a4d)
  - site_page: https://xiaofengshi.com/about (fetched 2026-08-29T09:47:42.765590+00:00, sha efbb4eb3e084)
- Data as of 2026-08-30T08:39:29.467469+00:00.