The goal is to find duplicate images via a beutiful web interface, where you can explore, select, move and delete single or a bunch of duplicate images from your computer.
Strong point of this software is pHash: an open source software library that implements several perceptual hashing algorithms, and provides a C-like API. It ignores the image size and file size and instead creates a hash based on the pixels of the image. This allows you to find duplicate pictures that have been rotated, have changed metadata, and slightly edited.
This code is a full refactor of this software, so thanks to @philipbl for the amazing work! :)
- Use a Virtual Environment that can be shared within the repository
- Use SQLite3 for the entire project to increase portabiliy
- Use Flask to create a beautiful, minimal and fast Web UI
- Write code as modular as possible
- Comment all the code
- Provide an accurate requirements.txt
- Provide multiple test's scripts
- Write a Wiki and a Quick Start guide
- Have fun!
- When (at least) a duplicate is found, display only one with a "ribbon" element from Semantic UI with a counter of how many duplicates software found.
- A click on "# Duplicates" display a modal where user can delete from local disk one or more duplicate images.
- In database save path and filename separately.
- Add path by drag and drop