Steady method of search of images in visual databases

Technical cybernetics. Information technology. Computer facilities


Аuthors

Sharonov A. V.*, Maximov N. А.**, Sincha D. P.

Moscow Aviation Institute (National Research University), 4, Volokolamskoe shosse, Moscow, А-80, GSP-3, 125993, Russia

*e-mail: anatoly.sharonov@yandex.ru
**e-mail: N-A-Maximov47@yandex.ru

Abstract

Considerable interest in unmanned aerial vehicles have recently increased the need for an effective system of pattern recognition [1]. Unlike other systems, pattern recognition (biometric identification systems, facial recognition systems, systems of identification documents, etc.), direct object, subject to classification (recognition) in this case, is the image itself, which, of course, can be distorted by various factors. At the heart of any system of pattern recognition are mainly two algorithms: an algorithm for forming a handle (index) of the image (i.e, an organized set of attributes of the image) and an algorithm for estimating the degree of similarity between two images in their descriptors. Quite promising variant of these algorithms is presented in this article.

Keywords:

pattern recognition, image features, the degree of similarity, search in image databases, image distortion, image classification


Download

mai.ru — informational site MAI

Copyright © 2000-2024 by MAI

Вход