dominhhai / Captcha Breaker
Licence: mit
High Accuracy Captcha Breaker with Tensorflow and Node.js
Stars: ✭ 76
Programming Languages
python
139335 projects - #7 most used programming language
Labels
Projects that are alternatives of or similar to Captcha Breaker
Rucaptcha
Captcha gem for Rails Application. No dependencies. No ImageMagick, No RMagick.
Stars: ✭ 607 (+698.68%)
Mutual labels: captcha
Hackcaptcha
provide api functions for captcha recognition.
Stars: ✭ 25 (-67.11%)
Mutual labels: captcha
Multicaptchabot
The best bot for collecting cryptocurrency from freebitco.in, freedoge.co.in and freenem.com 🚀🌔
Stars: ✭ 27 (-64.47%)
Mutual labels: captcha
Awesome Captcha
🔑 Curated list of awesome captcha libraries and crack tools.
Stars: ✭ 566 (+644.74%)
Mutual labels: captcha
Django Simple Captcha
Django Simple Captcha is an extremely simple, yet highly customizable Django application to add captcha images to any Django form.
Stars: ✭ 1,151 (+1414.47%)
Mutual labels: captcha
Phpauth
PHPAuth is a secure PHP Authentication class that easily integrates into any site.
Stars: ✭ 748 (+884.21%)
Mutual labels: captcha
Beelabrecaptcha2bundle
💻 Symfony bundle for Google Recaptcha2
Stars: ✭ 47 (-38.16%)
Mutual labels: captcha
Captcha Tensorflow
Image Captcha Solving Using TensorFlow and CNN Model. Accuracy 90%+
Stars: ✭ 660 (+768.42%)
Mutual labels: captcha
Challenge Bypass Extension
Privacy Pass: a privacy-enhancing protocol and browser extension.
Stars: ✭ 679 (+793.42%)
Mutual labels: captcha
Captchaimageview
Custom ImageView to generate captcha image.
Stars: ✭ 609 (+701.32%)
Mutual labels: captcha
Easycaptcha
Java图形验证码,支持gif、中文、算术等类型,可用于Java Web、JavaSE等项目。
Stars: ✭ 1,084 (+1326.32%)
Mutual labels: captcha
Recaptcha
reCAPTCHA = REcognize CAPTCHA: A Burp Suite Extender that recognize CAPTCHA and use for intruder payload 自动识别图形验证码并用于burp intruder爆破模块的插件
Stars: ✭ 596 (+684.21%)
Mutual labels: captcha
Invisible captcha
🍯 Unobtrusive and flexible spam protection for Rails apps
Stars: ✭ 851 (+1019.74%)
Mutual labels: captcha
Happy Captcha
Happy Captcha是一款易于使用的Java验证码软件包,旨在花最短的时间,最少的代码量,实现Web站点的验证码功能。Happy Captcha完全遵循Apache 2.0开源许可协议,你可以自由使用该软件,如您在使用Happy Captcha时发现软件的任何缺陷,欢迎随时与我联系。
Stars: ✭ 75 (-1.32%)
Mutual labels: captcha
Crapcha
Completely Ridiculous And Phony Captcha that Hassles for Amusement
Stars: ✭ 42 (-44.74%)
Mutual labels: captcha
Captcha Breaker
Build with Tensorflow (ConvNets) and Node.js 💪💪💪
E.x: Amazon Captcha (click image below to watch demo video)
Installation
Python packages
$ pip install -r requirements.txt
Node.js packages (Node.js user only)
$ npm i
Usage
1. Create train data
Prepare your training dataset
- Copy captcha images to
data/captcha
folder
|_data
|_captcha
|_ xss7.jpg
|_ tvu4.jpg
IMPORTANT: Note each image file is named with it's own solution.
That means that if an image is named A1bD3.jpg
, it corresponds to a captcha's whose solution is A1bD3
Build train data for model
Run src/create_train_data.py
will save your train data as data/captcha.npz
compressed file.
$ python src/create_train_data.py
The compressed data/captcha.npz
includes:
- Train Data (
x_train
,y_train
):80%
- Test Data (
x_test
,y_test
):20%
2. Train
Run src/train.py
to train the model with your own dataset.
$ python src/train.py
Take ☕️ or 🍵 while waiting!
3. Attack
Now, enjoy your war 🔥🔥🔥 😜😜😜
Python
$ python src/predict --fname YOUR_IMAGE_PATH_or_URL
Sample output:
loading image: data/captcha/captcha_2.jpg
load captcha classifier
predict for 1 char: `X` with probability: 99.956%
predict for 2 char: `I` with probability: 99.909%
predict for 3 char: `N` with probability: 99.556%
predict for 4 char: `C` with probability: 99.853%
predict for 5 char: `H` with probability: 99.949%
predict for 6 char: `A` with probability: 98.889%
Captcha: `XINCHA` with confident: `99.686%`
XINCHA
Node.js
const captchaPredict = require('src/predict')
captchaPredict(YOUR_IMAGE_PATH_or_URL)
.then(console.log)
.catch(console.error)
Sample output:
[
"loading image: data/captcha/captcha_2.jpg",
"load captcha classifier",
"predict for 1 char: `X` with probability: 99.956%",
"predict for 2 char: `I` with probability: 99.909%",
"predict for 3 char: `N` with probability: 99.556%",
"predict for 4 char: `C` with probability: 99.853%",
"predict for 5 char: `H` with probability: 99.949%",
"predict for 6 char: `A` with probability: 98.889%",
"Captcha: `XINCHA` with confident: `99.686%`",
"XINCHA"
]
Note that the project description data, including the texts, logos, images, and/or trademarks,
for each open source project belongs to its rightful owner.
If you wish to add or remove any projects, please contact us at [email protected].