Detekcija igrača u snimkama nogometnih utakmica

undergraduate thesis

• Title (english): Detection of Players in Football Game Video
• Author: Marko Kršić
• Mentor: Zoran Kalafatić (mentor)
• Committee member: Zoran Kalafatić (predsjednik povjerenstva)
• Committee member: Josip Šarić (član povjerenstva)
• Committee member: Marin Oršić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2019-07-12, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Analiza kretanja igrača pruža vrijedne informacije treneru i stručnom stožeru te je s obzirom na dosadašnji razvoj računalne tehnologije i dostupnost kamera visokih razlučivosti sasvim razumno olakšati, ubrzati i unaprijediti postupak ovakve analize primjenom računalnog vida. Osim što se izbjegava ponavljajuć i jednoličan posao, dolazi se do preciznijih i opširnijih informacija koje se lako mogu sažeti i sintetizirati. Ovakav pristup je vrlo elegantan jer ne zahtijeva opremu koja bi djelovala invazivno na samu igru ili igrače.
Za ostvarenje takvog sustava jedan od potrebnih koraka je implementacija postupka detekcije igrača. Pored raznih metoda detekcije u ovom je radu odabrana detekcija metodom oduzimanja pozadine koja je vrlo intuitivna i jednostavna te, sa svojim prednostima i nedostacima (npr. zahtjevom za stacionarnošću kamera), odlično ispunjava svoju ulogu u ovakvoj primjeni. Pri izvedbi ove metode su potrebne razne slikovne obrade te rješenja za probleme kao što je promjenjivost osvjetljenja terena. U ovom je radu opisana svojevrsna izvedba metode oduzimanja pozadine te je primijenjena na snimku nogometne utakmice na kojoj je potom i ocijenjena njena uspješnost.

Abstract (english)

The football players' movement analysis provides valuable information to the head coach and the rest of coaching staff. To further ease, speed up and improve the process of analysis it is reasonable to apply computer vision considering the modern development of computer technology and the availability of high-resolution cameras. Besides avoiding the repetitive and monotone work, the information acquired this way is more precise and in-depth while being simpler to summarize and synthesise. The reason why this approach is of such elegance is its lack of need for additional, potentially intrusive, equipment.
One of the steps to realization of such system is the implementation of player detection. In this study, amongst all the techniques of detection, the background subtraction is chosen due to its simplicity and intuitiveness while also fulfilling its role well besides having its requirements such as stationary cameras. For the realization of this method a variety of image processing is needed as well as solutions to problems such as the varying light conditions. In this study a detection via background subtraction is discussed, implemented and applied to a football game video as well as evaluated afterwards.

• Keywords: računalni vid; oduzimanje pozadine; detekcija nogometnih igrača
• Keywords (english): computer vision; background subtraction; football player detection
• Language: croatian
• URN:NBN: urn:nbn:hr:168:390087
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2020-01-07 19:28:50