Naslov
Core metamodel for content-based image retrieval

Autori
Ivašić-Kos, Marina ; Pavlić, Mile ; Poščić, Patrizia

Vrsta, podvrsta i kategorija rada
Radovi u zbornicima skupova, cjeloviti rad (in extenso), znanstveni

Izvornik
MIPRO 2008- 31st International convention on information and communication technology, electronics and microelectronics, Proceedings Vol. III, CTS & CIS / Bogunović, Nikola ; Ribarić, Slobodan - Rijeka : Hrvatska udruga za informacijsku i komunikacijsku tehnologiju, elektroniku i mikroelektroniku - MIPRO, 2008, 178-183

ISBN
978-953-233-038-0

Skup
MIPRO 2008 - 31st International convention on information and communication technology, electronics and microelectronics

Mjesto i datum
Opatija, Hrvatska, 26.05.2008. - 30.05.2008

Vrsta sudjelovanja
Predavanje

Vrsta recenzije
Međunarodna recenzija

Ključne riječi
content-based image retrieval; UML modeling; metamodel

Sažetak
Content-based image retrieval (CBIR) is a technique for retrieving images from a large data collection that match to a given query based, basically, on their own low-level visual content, such as colors, textures and shapes within the image. The content-based retrieval process usually involves generating signatures from extracted visual features and annotations of images, comparing these signatures to the signature of the query. Results are usually ranked in terms of how similar their signatures are to the query signature according to appropriate distance measure. The core concepts of the CBIR paradigm, i.e. visual feature extraction, signature definition with corresponding similarity metrics and retrieval techniques will be discussed and modeled using Unified Modeling Language (UML) modeling concepts. The aim of proposed UML core metamodel is to present multiple views of the system and to provide a formal basis for better understanding and improvements in CBIR systems development.

Izvorni jezik
Engleski

Znanstvena područja
Informacijske i komunikacijske znanosti

POVEZANOST RADA