03lenio / WebCaptchaScraper

WebCaptchaScraper is a program to scrape captchas from web.de as base for machine learning

WebCaptchaScraper

Introduction

This program is for everyone who wants to auto-solve captchas using machine learning, this program can generate you a database of 1 to ∞ captcha samples, the standard is web.de a german mail provider but you can easily change this in code :)

Code Samples

code for the scrape method:

# create scraped folder if not exist
if not os.path.exists("scraped"):
    os.mkdir("scraped")
time.sleep(2)
# this will call the global captcha count to count how many captchas you already have
global captchaCount
captchaCount += 1
# here selenium finds the captcha image via its id and then screenshots it
try:
    driver.find_element_by_id("captchaImage").screenshot("scraped/captcha" + str(captchaCount) + ".png")
except:
    print("Something went wrong while trying to scrape, retrying...")
    grabCaptchas()
print("Successfully grabbed captcha no." + str(captchaCount))
global grabbed
grabbed += 1
# this will refresh the site so another captcha can be scraped
driver.refresh()

Installation

The easiest way of installing it, is using the release which will be updated every once in a while, but if you want to scrape captchas from another site than web.de you'll have to install Selenium with the command: pip install selenium, after downloading this you can easily change your script and execute itwithin any shell!