crta
Hrvatska znanstvena Sekcija img
bibliografija
3 gif
 Naslovna
 O projektu
 FAQ
 Kontakt
4 gif
Pregledavanje radova
Jednostavno pretraživanje
Napredno pretraživanje
Skupni podaci
Upis novih radova
Upute
Ispravci prijavljenih radova
Ostale bibliografije
Slični projekti
 Bibliografske baze podataka

Pregled bibliografske jedinice broj: 883825

Časopis

Autori: Braović, Maja; Stipaničev, Darko; Krstinić, Damir
Naslov: Cogent Confabulation based Expert System for Segmentation and Classification of Natural Landscape Images
Izvornik: Advances in Electrical and Computer Engineering (1582-7445) 17 (2017), 2; 85-94
Vrsta rada: članak
Ključne riječi: Cogent confabulation, color spaces, image segmentation, image classi cation, expert systems, information fusion, natural landscape images
Sažetak:
Ever since there has been an increase in the number of automatic wildfire monitoring and surveillance systems in the last few years, natural landscape images have been of great importance. In this paper we propose an expert system for fast segmentation and classification of regions on natural landscape images that is suitable for real-time applications. We focus primarily on Mediterranean landscape images since the Mediterranean area and areas with similar climate are the ones most associated with high wildfire risk. The proposed expert system is based on cogent confabulation theory and knowledge bases that contain information about local and global features, optimal color spaces suitable for classification of certain regions, and context of each class. The obtained results indicate that the proposed expert system significantly outperforms well-known classifiers that it was compared against in both accuracy and speed, and that it is effective and efficient for real-time applications. Additionally, we present a FESB MLID dataset on which we conducted our research and that we made publicly available.
Izvorni jezik: ENG
Rad je citiran u
bazama podataka:
Web of Science: Science Citation
Web of Science: Science Citation Index Expanded
Scopus
Kategorija: Znanstveni
Znanstvena područja:
Računarstvo
URL Internet adrese: http://www.aece.ro/abstractplus.php?year=2017&number=2&article=12
Broj citata:
Altmetric:
DOI: 10.4316/AECE.2017.02012
URL cjelovitog teksta: http://www.aece.ro/abstractplus.php?year=2017&number=2&article=12
Časopis izlazi u samo elektroničkom izdanju: NE
Google Scholar: Cogent Confabulation based Expert System for Segmentation and Classification of Natural Landscape Images
Upisao u CROSBI: Darko Stipaničev (Darko.Stipanicev@fesb.hr), 4. Srp. 2017. u 09:25 sati



Verzija za printanje   za tiskati


upomoc
foot_4