# kerlomz/captcha_trainer

[验证码识别-训练] This project is based on CNN/ResNet/DenseNet+GRU/LSTM+CTC/CrossEntropy to realize verification code identification. This project is only for training the model.

Repository: https://github.com/kerlomz/captcha_trainer
Canonical: https://ross.abutalabs.com/products/captcha_trainer
Language: Python
License: Apache-2.0
License Family: permissive
Topics: captcha-recognition, tensorflow-tutorials, ocr, tensorflow
Last push: 2025-11-09T11:37:45+00:00

## Health v2 (maintenance only)
Score: 55/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 51, release rhythm 35, longevity 100
- inputs: {"age_days": 2938, "days_push": 297, "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 3213, forks 823 (observed 2026-08-28T04:07:49.411997+00:00)

## What it is
A deep learning training tool for image CAPTCHA recognition built on TensorFlow, using CNN/ResNet/DenseNet backbones with GRU/LSTM recurrent layers and CTC/CrossEntropy decoding. It offers a visual, code-free workflow for training OCR models that can be exported and deployed via a companion serving platform.

## Use cases
- train a model to recognize image captchas
- build an OCR model for distorted text images without coding
- train text recognition models with CTC loss
- create image classification models from labeled samples
- recognize captchas with overlapping or rotated characters

## When to avoid
- you need a ready-made inference/serving solution rather than training (use the companion deployment project)
- you work outside TensorFlow 1.x ecosystems or need PyTorch
- you need general-purpose document OCR rather than short text/captcha recognition

## Facets
- artifact type: application
- maturity: active
- function: machine-learning, deep-learning, ocr, image-processing, llm-training
- domain: machine-learning, computer-vision, image-processing, developer-tools
- platform: windows, python, cross-platform
- tags: captcha-recognition, tensorflow, ocr-training, ctc-loss, model-training, gui, no-code, gpu

## Member repositories
- kerlomz/captcha_trainer (main) score 55

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:49.411997+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 2026-08-30T07:24:42.099898+00:00, confidence not recorded.
  - readme: https://github.com/kerlomz/captcha_trainer (fetched 2026-08-28T04:07:49.411997+00:00, sha a295291de646)
- Data as of 2026-08-30T08:39:29.467469+00:00.