Shutter by Akis Papadopoulos | New in the App Gallery

Welcome to the OpenShift Developer Spotlight where we get to know the members of the OpenShift community a little better and show off their skills as developers. Also check out their applications developed on OpenShift in our App Gallery. Interested in being featured? Apply here or view past entries.

What inspired you to be a developer?

Being a developer is like walking in a hidden, rough path full of challenges where you're learning new technologies, creating things from scratch, and sharing your knowledge to the community.

Why did you choose OpenShift as your hosting platform?

I think OpenShift is more than just a hosting service, it is actually a reliable platform in which you can invest both time and money in order to create, deploy and scale your apps so easily and fast.

What advantages does OpenShift give you that other platforms don't?

The use of the many cartridges you can find in the marketplace, the scalability factor you can get so instantly and fast, and also the option to work in the free tier which helps in small teams.

Tell us more about your application currently hosted on OpenShift:

• Name: Shutter
• What does it do? Shutter is an experimental image search engine by content (cbir), a web application searching similar images/photos of famous landmarks around the world, indexed already in third party web sites like flickr.com and 500px.com. This is a personal project and we're running on a short list of only 17 indexed landmarks like the Brandeburg Gate in Berlin, the Parthenon in Athens, the Ancient Theater of Epidaurus, the White Tower of Thessaloniki, the London Eye, the Big Ben in London and so on.
• What technologies were used to create your app? Java, Tomcat, JSP/Servlets and PostgreSQL.
• What motivated you to create this application and what problems does it solve? A few years ago, while completing my Masters degree in the school of Computer Science, I worked for a full semester focusing on information retrieval theory on visual content. So I decided to use the theoretical background in order to create an experimental service to search similar images regarding only the visual content ignoring any textual metadata. It works in a simple way, a user uploads an image of a famous landmark like Brandenburg Gate in Berlin, then waits to get back a list of similar images taken by photographers and travelers stored within services like flickr and 500px.