LIRE: Lucene Image Retrieval
LIRE is a Java library that provides a simple way to retrieve images and photos based on color and texture characteristics. LIRE creates a Lucene index of image features for content based image retrieval (CBIR) using local and global state-of-the-art methods. Easy to use methods for searching the index and result browsing are provided. Best of all: it's all open source.
LIRE is successfully used at the WIPO, a United Nations Agency, to search in millions of trademark images and the Danish National Police to find similar scenes and to detect near duplicates.
Features OverviewLIRE works on platforms supporting Java 1.7 and has been tested on Windows and Linux. Test it on a million images with the web demo.
LIRE comes without secrets. All the state of the art algorithms are available in source.
State of the Art
Well known and well performing local and global features are integrated.
Using approximate search LIRE scales well to millions of images.
Take control over every step in the process, tinker with extraction, indexing and search.
You need a search server? There is a plugin for the popular Apache Solr search server.
Fresh & new algorithms from the visual information retrieval community.
GNU GPL License
LIRE is published under the viral OSS Licene, the GNU Public License. Therefore, it will stay free.
Hosted on Github
LIRE source is hosted and managed on Github. Check out revisions, submit patches and more.
Anything you need to know? Ask on the LIRE mailing list. We are happy to help you!
Lots of examples
Developers will find everything to start with at the developers' docs. Even sample applications for common tasks are available.
It even has a book
"Visual Information Retrieval using Java and LIRE", written by Mathias Lux and Oge Marques is available on Amazon for a little more than 10$
De-Facto Reference Software
For many of the global descriptors LIRE is the de-facto reference implementation, also in research.