Image Search

An Electron-based desktop application for scraping images from websites and searching for similar images using perceptual hashing.

Features

• Web Scraping: Scrape images from websites using Puppeteer
• Image Search: Find similar images using perceptual image hashing
• Batch Processing: Process multiple pages automatically
• Anti-Detection: Uses stealth plugins to avoid detection

Requirements

• Node.js (v14 or higher)
• npm

Installation

npm install

Usage

1. Start the application:

npm start

2. Scrape Images:

   • Enter a URL in the URL input field
   • Click "Scrape Images" to start scraping
   • Images will be saved to the temp_images folder

3. Search for Similar Images:

   • Select a reference image using the file input
   • Click "Search" to find similar images in the scraped collection
   • Results are displayed sorted by similarity

4. View Temp Files:

   • Click "Temp File" to open the temp_images folder

Dependencies

• electron - Desktop application framework
• puppeteer-extra - Web scraping with stealth features
• image-hash - Perceptual image hashing for similarity detection
• axios - HTTP client for downloading images

Project Structure

image-search/
├── main.js          # Main Electron process
├── preload.js       # IPC bridge between renderer and main process
├── renderer.js      # Frontend logic
├── index.html       # UI markup
├── main.css         # Styling
└── temp_images/     # Downloaded images directory (auto-created)

Notes

• Images are temporarily stored in temp_images/ directory
• The similarity threshold for image matching is set to 50 (adjustable in main.js)
• The scraper is configured for a specific domain keyword filter