20.500.12556/DKUM-61970 UPORABA METOD STROJNEGA UČENJA ZA DOLOČANJE DOMINANTNE BARVE NA DIGITALNI FOTOGRAFIJI USING MACHINE LEARNING METHODS FOR DETERMINING DOMINANT COLOR ON DIGITAL PHOTOGRAPHY V diplomskem delu smo podrobno raziskali algoritem k-means, ki v kontekstu strojnega učenja spada v skupino algoritmov nenadzorovanega učenja. Proučili smo probleme in možnost iskanja dominantnih barv na digitalni fotografiji z uporabo omenjenega algoritma. Raziskali smo različne barvne modele in iskanje barvnih harmonij. Izdelali smo aplikacijo, ki omogoča fotografiranje (npr. oblačil), iskanje dominantnih barv in določanje različnih barvnih harmonij. In this thesis we carried out a detailed research of k-means algorithm, which in the context of machine learning is classified into the group of unsupervised learning algorithms. We examined the possibilities for determining dominant color in the digital photography using the k-means algorithm. We also researched various color models and how to find color harmonies. In the end, an application was implemented with which the user can take a picture (e.g. of clothing), find dominant colors and determine the color harmonies. strojno učenje k-means barvni model dominantna barva barvne harmonije machine learning k-means color model dominant color color harmonies true true false D. Šnajder Slovenski jezik Angleški jezik Diplomsko delo 2016-08-19 22:34:38 2016-09-06 12:54:27 2022-06-28 03:07:10 2019-07-11 12:39:18 2016 0 [Maribor 0 0000-00-00 NiDoloceno NiDoloceno NiDoloceno 0000-00-00 0000-00-00 0000-00-00 004.85.021(043.2) 19844374 URN:SI:UM:DK:LV7I4SFF UN_Snajder_Dario_2016.pdf UN_Snajder_Dario_2016.pdf 1 F8FDE581B6FE08EB57DF168F3202CB01 3818dcab5967840b63f7b5c1ad7fb949a76164b3cbf9fad973ebef2c5189da0e 0cd55328-7c0d-11eb-bb7a-00155d0001ca https://dk.um.si/Dokument.php?lang=slv&id=99145 Fakulteta za elektrotehniko, računalništvo in informatiko 0 0 0