Search by Image Engine using Local Feature Detectors

Abstract

Nowadays, modern big data warehouses require the development of effective management algorithms. For large image storages first of all it's important to develop effective algorithms for comparing and searching a similar image from an image, taking into account their possible geometric transformations. In this regard, one of the most promising is the approach based on the use of invariant Local Feature Detectors. The work is carried out experimental research of such detectors’ effectiveness for Search by Image Engine in large image storages.