B.A.R.I.S.T.A. (WIP)

Basic-Arse Reverse Image Search Tool (Amazing)

BARISTA is my attempt at a reverse image search tool that is powered by hashing. If you remember from class, hashing algorithms like SHA256 will produce a completely unique hash for each image and file, with huge differences in the resulting hash even if only a small amount of file/image data has been changed. Algorithms such as Perceptual Hash and Difference Hash do the opposite, they will still give an identical (or similar) hash if small amounts of image data have been changed. We can take advantage of this to discover other files that have some similarity.

BARISTA is a self-contained Javascript web application that handles a persistent SQLite database full of image hash data that is accessible through an API and through a browser GUI.

Deployment

1. Download the repository somewhere.
2. cd into the frontend directory and run npm install. Wait for prerequisites to download, and then do npm build.
3. cd into the backend directory and run npm install. Wait for prerequisites to download.
4. Do npm start while in the backend directory to launch BARISTA. How to populate the database (and how to set it up to work with your own web scraper) is covered under the API subheading.

Usage

Access the service in your browser by visiting localhost:5000. You can use port forwarding to also make this service accessible over the internet.

API

Sample Data

For the purposes of demonstration and testing, I've included some sample data which is contained in the "companion" folder. This data needs to be fed to BARISTA's API to be useable.

• metmuseum_images.txt contains 7416 image links (one per line) from the Met Museum.
• populator.py is a very simple Python script to feed image links to the database's API.
• The "test" folder contains some images from the Met Museum's website, as well as some altered variants.

First, make sure Barista's backend is running. Then just run the Python script to begin automatically feeding the data into Barista.