Prepoznavanje objekata pristupanjem sadržaju digitalne slike

master's thesis

• Title (english): Object recognition in content-based digital image retrieval
• Author: Luka Antičević
• Mentor: Dragan Jevtić (mentor)
• Committee member: Dragan Jevtić (predsjednik povjerenstva)
• Committee member: Mirko Randić (član povjerenstva)
• Committee member: Miljenko Mikuc (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

U okviru ovog rada, korištenjem OpenCV knjižnice i programskog jezika Python, izrađen je sustav za pronalaženje i prepoznavanje prometnih znakova koji se temelji na modelu umjetne neuronske mreže. Napravljena je analiza modela i njegove sposobnosti indentifikacije objekata za različite konfiguracije mreže. Objašnjeni su modeli zbirke svojstava i algoritam SURF koji se koriste prilikom procesa pronalaženja znakova. Na kraju su prezentirani eksperimentalni rezultati dobiveni na skupu od 100 slika... More za testiranje. Less

Abstract (english)

The main goal of this thesis is building a system for detecting and recognizing traffic signs using OpenCV library and Python programming language. System is based on the model of artificial neural network. Analysis of the recognition rate in relation to different network configurations was conducted in the scope of the thesis. A detailed description of Bag of Features model and SURF algorithm are given. Finally, experimental results obtained from set of 100 pictures are presented.

• Keywords: prepoznavanje objekata; umjetne neuronske mreže; algoritam SURF; zbirka svojstava; OpenCV
• Keywords (english): object recognition; artificial neural networks; SURF algorithm; bag of features; OpenCV
• Language: croatian
• URN:NBN: urn:nbn:hr:168:278773
• Study programme: Title: Information and Communication Technology; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije (magistar/magistra inženjer/inženjerka informacijske i komunikacijske tehnologije)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2018-02-06 07:37:01