Usporedba metoda računalnog vida i metoda dubokog učenja za prepoznavanje i klasifikaciju objekata u slikama

master's thesis

• Title (english): Comparison of computer vision and deep learning methods for objects recognition and classification in images
• Author: Hrvoje Ditrih
• Mentor: Sonja Grgić (mentor)
• Committee member: Sonja Grgić (predsjednik povjerenstva)
• Committee member: Tomislav Kos (član povjerenstva)
• Committee member: Marko Bosiljevac (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2021-09-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Set je igra s kartama u stvarnom vremenu koja zahtjeva vještine vizualizacije, logičkog razmišljanja te sposobnost koncentracije. Igranje Set-a može biti intelektualno zahtjevno za ljude, dok računala mogu igrati Set ljudima nezamislivom brzinom. S druge strane, prepoznavanje karata igre Set ljudima je intuitivno i mogu to raditi bez svjesnog razmišljanja. Kako je teško formalno opisati način kojim ljudi prepoznavaju objekte, tako je i iznimno zahtjevno razviti algoritme pomoću kojih će računala prepoznavati objekte sa slike. U ovom radu predstavljeni su postupci prepoznavanja i klasifikacije karata igre Set pomoću klasičnih metoda računalnog vida i metoda dubokog učenja te je predstavljen sustav SetFinder koji igra Set u stvarnom vremenu.

Abstract (english)

Set is a card game in real-time which requires the skill of visualization, logical thinking and the ability to focus. Playing Set can be intellectually challenging for people but computers can play Set at unimaginable speeds. On the other hand, recognizing the cards of the game is intuitive for people and they can do it without conscious thinking. Computers are bad at object recognition because it is really hard to formally describe the process of how people recognize objects. This thesis presents procedures of both classic computer vision and deep learning for recognizing and classifying the cards of the game Set. Furthermore, it presents the SetFinder system which can play Set in real-time.

• Keywords: računalni vid; duboko učenje; igra Set; prepoznavanje i klasifikacija karata u slikama
• Keywords (english): computer vision; deep learning; Set card game; detection and classification of cards in images
• Language: croatian
• URN:NBN: urn:nbn:hr:168:681670
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-01-13 14:48:39