← all repositories

996refuse/zheye

A Chinese captcha recognition program that identifies inverted Chinese characters in Zhihu login captchas using CNNs and Gaussian mixture models.

793 stars Python Computer Vision
zheye
Not currently ranked — collecting fresh signals.
star history

This project uses convolutional neural networks combined with Gaussian mixture models to recognize and locate upside-down Chinese characters in Zhihu captcha images. The model is trained on captcha samples and can evaluate new images to identify character positions. It requires PyTorch, scikit-learn, and standard image processing libraries.

Frequently asked

What is 996refuse/zheye?
A Chinese captcha recognition program that identifies inverted Chinese characters in Zhihu login captchas using CNNs and Gaussian mixture models.
Is zheye open source?
Yes — 996refuse/zheye is an open-source project tracked on heatdrop.
What language is zheye written in?
996refuse/zheye is primarily written in Python.
How popular is zheye?
996refuse/zheye has 793 stars on GitHub.
Where can I find zheye?
996refuse/zheye is on GitHub at https://github.com/996refuse/zheye.

heatdrop uses Google Analytics to see which pages get read — nothing else. Your call. How we handle data.