sgbalogh / mass-vision

Mongo-backed Rails photo archive application

Mass-Vision

A Ruby on Rails photo archive application

Introduction

This application, currenly called Mass-Vision (until I think of something more clever), was built to support large batch image uploads, and to scrape as much meaning as possible from photographic metadata and location data, where it exists.

It relies on MongoDB for the database backend, Apache Tika for metadata scraping, RIIIF for an IIIF-compatible image service, and Delayed Job for asynchronous upload processing.

This project is very much incomplete, though cloning the master branch at any time (and installing all software prerequisites) should get you a working application.

Installation

I'm running this application on OS X 10.11, though it should be easy to deploy on Linux as well.

To install MongoDB on OS X:

brew update
brew install mongodb

Install Apache Tika, for metadata extraction:

brew install tika

Install ImageMagick, a dependency for RIIIF:

brew install imagemagick --with-ghostscript --with-tiff --with-jp2

Finally, clone the repo:

cd ~
git clone https://github.com/sgbalogh/mass-vision.git

From there, you can run the database migrations (rake db:migrate), and try booting up the application (rails s -b 0.0.0.0). To start a Delayed Job worker, run the rake task rake jobs:work, and hit Ctrl-C when you are ready to stop the worker.

Screenshots

Above: a view for an image with lat/lon geolocation