# dee1024/pytorch-captcha-recognition

基于CNN训练的一套 "端到端" 的验证码识别模型，使用深度学习+训练数据+大量计算力，纯数字识别率高达 99.99%，数字+字母识别率 96%

Repository: https://github.com/dee1024/pytorch-captcha-recognition
Canonical: https://ross.abutalabs.com/products/pytorch-captcha-recognition
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2020-01-09T10:47:29+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3081, "days_push": 2428, "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 1220, forks 327 (observed 2026-08-28T04:04:01.862609+00:00)

## What it is
A PyTorch-based end-to-end CNN model for recognizing captchas (digits and letters) without manual image preprocessing, using self-generated training data via ImageCaptcha. It serves primarily as an educational project demonstrating deep learning OCR workflows.

## Use cases
- recognize captcha images with deep learning
- train a CNN to read digits and letters from images
- learn end-to-end OCR with pytorch
- generate labeled captcha training data
- build a multi-label character classification model

## When to choose
- you want to learn CNN-based OCR with PyTorch
- you need to recognize simple generated captchas of digits and letters
- you want a complete train/test/predict pipeline example

## When to avoid
- you need production-grade OCR for complex or noisy captchas
- you require actively maintained software with recent updates
- you need support for arbitrary text, distorted, or adversarial captchas

## Facets
- artifact type: learning-resource
- maturity: maintenance
- function: machine-learning, deep-learning, ocr, image-processing, computer-vision
- domain: deep-learning, computer-vision, image-processing, machine-learning, tutorials
- platform: python, cross-platform
- tags: captcha-recognition, pytorch, cnn, ocr, end-to-end, image-classification

## Member repositories
- dee1024/pytorch-captcha-recognition (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:04:01.862609+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 2026-08-30T06:15:57.139364+00:00, confidence not recorded.
  - readme: https://github.com/dee1024/pytorch-captcha-recognition (fetched 2026-08-28T04:04:01.862609+00:00, sha 1bac96034b89)
- Data as of 2026-08-30T08:39:29.467469+00:00.