ddddocr

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

14665 stars · 2336 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

DdddOcr is a Python library for offline, local recognition of various CAPTCHA types, including alphanumeric, Chinese character, and slider CAPTCHAs, using deep learning models trained on large volumes of generated data. It also supports object detection in images and importing custom-trained models, with a minimal-dependency, easy-to-use API.

Use cases

• recognize text from captcha images
• solve alphanumeric captchas offline in python
• detect slider captcha gap position
• read chinese character captchas
• detect text regions in images
• use a custom trained ocr model for captchas
• batch recognize captcha images with multithreading

When to choose

• you need offline, local captcha recognition without paid APIs
• you want a simple pip-installable OCR SDK with minimal setup
• you need to handle slider or Chinese captchas
• you want to plug in your own trained captcha models

When to avoid

• you need to solve reCAPTCHA, hCaptcha, or other interactive commercial captchas
• you need general-purpose document OCR rather than short captcha strings
• you require 32-bit Windows or Linux support
• you need a hosted API service with managed scaling

Facets

library · maturity active

ocr computer-vision image-processing machine-learning computer-vision image-processing developer-tools python windows cross-platform captcha-recognition captcha-solving slider-captcha object-detection offline-ocr deep-learning automation linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="sml2h3/ddddocr")