Detekcija i praćenje igrača u snimkama nogometnih utakmica

master's thesis

• Title (english): Detection and Tracking of Players in Football Game Video
• Author: Karlo Dumbović
• Mentor: Zoran Kalafatić (mentor)
• Committee member: Zoran Kalafatić (predsjednik povjerenstva)
• Committee member: Siniša Šegvić (član povjerenstva)
• Committee member: Tomislav Hrkać (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2016-09-27, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

U ovom diplomskom radu opisan je sustav za detekciju i praćenje igrača u snimkama nogometnih utakmica koji koristi tehnike računalnog vida. Sustav je razvijen koristeći snimke snimljene nepokretnom kamerom, ali su predloženi efikasni alternativni pristupi koji su iskoristivi uz korištenje pokretne kamere. Tri okosnice rada su detekcija terena, detekcija igrača i praćenje igrača, od kojih uspješnost svake ovisi o uspješnosti onih prethodno navedenih. Za razvoj sustava korišteni su brojni računarski i matematički algoritmi i postupci, poput čestičnog i Kalmanovog filtra, k-d stabla, algoritma traženja minimalnog pokrivanja bridova u težinskom grafu, Houghove transformacije, te algoritma k-srednjih vrijednosti. Sustav je razvijen u programskom jeziku C++ uz korištenje biblioteke OpenCV.

Abstract (english)

The focus of this master thesis is developing a system for detection and tracking of players in video recordings of football matches using computer vision techniques. The system is built to work on recordings that were filmed with a stationary camera, but there are also some useful approaches that would work well with moving camera described. Three main parts of the thesis are field detection, player detection and player tracking, among which success of each one of them depends on the gain achieved by the ones previously listed. While building the system there were many math and computer science algorithms and procedures used, such as particle and Kalman filters, k-d trees, an algorithm to find minimal edge cover in weighted graph, Hough transform and k-means clustering. The system is built in programming language C++ using computer vision library OpenCV.

• Keywords: analiza nogometne igre; računalni vid; detekcija igrača; praćenje igrača; detekcija nogometnog terena
• Keywords (english): football game analysis; computer vision; player detection; player tracking; football field detection
• Language: croatian
• URN:NBN: urn:nbn:hr:168:887111
• 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: 2018-02-06 08:36:59