blkstone / simple-decaptcha

using OpenCV and tesseract to recognize CAPTCHA

using OpenCV and tesseract to recognize CAPTCHA

本项目是对验证码图片进行少量预处理，之后利用tesseract识别。 实验性项目，功能还非常不完善。

tesseract 配置方法

http://blkstone.github.io/2015/12/16/tessercat-ocr/

值得深入研究的参考项目与资料

PWNTCHA - Pretend We're Not a Turing Computer but a Human Antagonist http://caca.zoy.org/wiki/PWNtcha http://caca.zoy.org/browser/pwntcha

Google algorithm busts CAPTCHA with 99.8 percent accuracy http://www.zdnet.com/article/google-algorithm-busts-captcha-with-99-8-percent-accuracy/

谷歌CAPTCHA hacking http://www.cnbeta.com/articles/490545.htm 从事这项研究的三名专家分别是Suphanee Sivakorn、Jason Polakis以及Angelos D. Keromytis。他们的论文名为《I Am Robot: (Deep) Learning to Break Semantic Image CAPTCHAs》——I Am Robot正好跟CAPTCHA运行时，要用户打勾的“I'm not a robot”相对，当前已经公布在了哥伦比亚大学，计算机科学系的网站上。这份研究报告也公布在了Black Hat Asia 2016网站上。

验证码的未来：扒一扒reCAPTCHA的那些事 http://www.freebuf.com/news/special/57496.html

阿里云誉反欺诈 验证码对抗之路及现有验证机制介绍 http://www.freebuf.com/articles/security-management/105890.html

PESIT decaptcha http://www.slideshare.net/guneshwor/decaptcha-31651647

常见验证码的弱点与验证码识别 - insight-labs http://drops.wooyun.org/tips/141

Python识别网站验证码 http://drops.wooyun.org/tips/6313

yzm_decode https://github.com/payonesmile/yzm_decode

图像验证码和大规模图像识别技术 http://www.infoq.com/cn/articles/CAPTCHA-image-recognition

端到端的OCR：基于CNN的实现 http://blog.xlvector.net/2016-05/mxnet-ocr-cnn/

针对极验滑动式验证码的处理 精易论坛

ReBreakCaptcha：利用谷歌来破解谷歌的验证码 http://www.360zhijia.com/360anquanke/179207.html https://east-ee.com/2017/02/28/rebreakcaptcha-breaking-googles-recaptcha-v2-using-google/ PoC https://github.com/eastee/rebreakcaptcha

欺骗反作弊系统

Explaining and Harnessing Adversarial Examples 如何欺骗神经网络，让它把熊猫识别为秃鹫 http://mp.weixin.qq.com/s?__biz=MjM5MTQzNzU2NA==&mid=401623809&idx=1&sn=ed739498ae5a55e86e16dbf3676e20e9

Breaking Linear Classifiers on ImageNet Andrej Karpathy

这也许将是围棋AI的最后死穴 http://www.guokr.com/article/441154/ Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images

更多信息

github 搜索关键词 decaptcha

致谢

本项目部分参考于，感谢 JetMuffin https://github.com/JetMuffin/CodeAnalyzer