AutoCrawler is a Python-based image crawling tool designed to automatically download large numbers of images from search engines using automated browser interaction. It uses Selenium and a Chrome browser driver to navigate image search pages and collect image sources based on keywords provided by the user. AutoCrawler supports multiprocess and multithreaded downloading, which allows it to retrieve images faster by running several tasks simultaneously. Users provide search terms through a simple keyword file, and the crawler organizes downloaded images into directories for each keyword. It can download either thumbnails or full resolution images and supports multiple image formats such as JPG, GIF, and PNG. It also includes configuration options such as headless mode, download limits, proxy usage, and thread count to customize crawling behavior.
Features
- Multiprocess and multithreaded image downloading for improved speed
- Keyword-based image crawling using a simple configuration file
- Option to download full resolution images instead of thumbnails
- Headless browser mode for running without a graphical interface
- Proxy list support for distributing requests across multiple proxies
- Data imbalance detection to identify categories with fewer images