A multithreaded Google Images scraper without Chromium. Only requires the python standard library, requests, and a few helper libraries. Aimed at being cross-platform, with a preference towards linux.

A Note:

Scraping too aggressively can create large server loads, and lead to 503 errors. I'm sure this is less of an issue for Google, but please be considerate. Thanks :)

If you like the project give me a Star ⭐

As of now, there's no packaging or runner scripts. You have to clone the repo, manually install the dependencies, and run main with python.

git clone https://github.com/talleyhoe/google-image-scraper.git

pip install -r requirements.txt

$ -> python src/main.py [keyword]
optional arguments:
 --count, -c How many images to try to scrape ( < 500 usually works )
 --directory, -d What directory to save the images in 
 (Default is ~/Downloads/[keyword]))
 --threads, -t How many threads to scrape with
 (Default is single threaded)

• Proper packaging
• Extensibility to scrape other sites
  • May not implement for a while unless there is sufficient interest