Msc Semester Project
Title
JPSearch: Search engine for image databases
Candidate
Murielle Ange Tiambo
Supervisor
Prof. Touradj Ebrahimi
Assistant: Dr. Frédéric Dufaux
Expected
2008, summer semester
Place
EPFL
Description
Search engines are very important in order to be able to find relevant information in huge databases. Up to now, most developments have been on search of text documents (e.g. Google to search web pages). However, there is a growing interest in technologies for searching multimedia documents. To answer this demand, JPEG has started a new effort in order to standardize a norm, JPSearch, for the indexing and retrieval of still images.
More specifically, the goal of this project is to develop a prototype of a search engine to index and retrieve still images. The query is based on image similarity.
Precisely, the following tasks have to be preformed:
- Extract from the image a dominant color descriptor based on color histogram
- Extract from the image a texture descriptor based on Gabor wavelet transform
- Extract these color and texture descriptors on subsets of the image instead of the whole image
- Develop a search and retrieval system, taking a query image as input, and returning similar images from a database ranked by similarity