MgArcher/Text_select_captcha

Health v2 · maintenance only

Flags: no_releases no_license

Adoption not part of the score

1656 stars · 523 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A PyTorch-based deep learning system that recognizes click-based (text-select) CAPTCHAs by detecting and ordering Chinese character positions using YOLO object detection and a Siamese matching network. It provides a Python API, CLI demos, and a FastAPI-based REST service returning click coordinates.

Use cases

• recognize text-select click captchas from images
• solve match-the-characters captcha puzzles automatically
• serve captcha recognition via a REST API
• train a custom captcha recognition model with a small dataset
• integrate captcha solving into browser automation scripts

When to choose

• you need to recognize Chinese text-click captchas with high accuracy and low latency
• you have only a few hundred labeled captcha images for training
• you want a self-hosted captcha recognition service on low-resource hardware

When to avoid

• you need to solve non-click captchas like reCAPTCHA or hCaptcha
• you require a commercially licensed library (no license is provided)
• you cannot use PyTorch or GPU/CPU inference in your environment

Facets

library · maturity active

machine-learning computer-vision ocr http-server deep-learning computer-vision machine-learning crawlers python cross-platform windows captcha-recognition text-select-captcha yolo siamese-network pytorch captcha-solving automation linux macos

Member repositories

For agents

markdown · JSON · MCP: product_card(name="MgArcher/Text_select_captcha")