prikaz prve stranice dokumenta Usporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka
  Download
PDF 2.67 MB
master's thesis
Usporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka
2016. urn:nbn:hr:200:388356
Pul, Matija
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Communications
Chair of Radiocommunications and Telecommunications

Cite this document

Pul, M. (2016). Usporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:388356

Pul, Matija. "Usporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2016. https://urn.nsk.hr/urn:nbn:hr:200:388356

Pul, Matija. "Usporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2016. https://urn.nsk.hr/urn:nbn:hr:200:388356

Pul, M. (2016). 'Usporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 27 December 2024, https://urn.nsk.hr/urn:nbn:hr:200:388356

Pul M. Usporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2016 [cited 2024 December 27] Available at: https://urn.nsk.hr/urn:nbn:hr:200:388356

M. Pul, "Usporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2016. Available at: https://urn.nsk.hr/urn:nbn:hr:200:388356

Please login to the repository to save this object to your list.
Metadata
TitleUsporedba dinamičnih slika zasnovana na izdvajanju i usporedbi ključnih točaka
Title (english)Dynamic picture comparison based on extraction and comparison of key points
AuthorMatija Pul
MentorMario Vranješ (mentor)
Committee memberMario Vranješ (predsjednik povjerenstva)
Committee memberRatko Grbić (član povjerenstva)
Committee memberMarijan Herceg (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Communications)
(Chair of Radiocommunications and Telecommunications)
Osijek
Defense date and country2016-09-29, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Telecommunications and Informatics
Abstract
Tema ovog diplomskog rada je problem usporedbe dinamičnih slika. Dinamične slike su slike sa stalnim kontekstom čiji je sadržaj promjenjiv u vremenu. Standardni način računalne usporedbe slika zasnovan na usporedbi elemenata slike uspoređuje sadržaj slike te se zbog toga ne može primijeniti za usporedbu dinamičnih slika. U sklopu ovog diplomskog rada izrađen je algoritam za detekciju objekata zasnovan na izdvajanju i uspoređivanju ključnih točaka. Algoritam koristi ORB algoritam za detekciju ključnih točaka. Glavni problem koji se pojavljuje je pogrešna detekcija ključnih točaka. Pogrešna detekcija se događa kada algoritam usporedi referentnu ključnu točku s testnom i dobije pozitivan rezultat, iako se radi o pogrešnoj točki (najčešće kod teksta kada se ista slova ili riječi pojave na više mjesta). Taj problem je riješen definiranjem objekata gdje se svakoj ključnoj točki dodjeljuje objekt (grupa) kojem pripada te se filtriraju sva poklapanja s pogrešnim objektima. Algoritmu je testirana otpornost na smetnje (rotaciju, skaliranje, pojavu čestica i translaciju) te učinkovitost rada na stvarnim sekvencama.
Abstract (english)
This Master's Thesis addresses the problem of dynamic image comparison. Dynamic images are images with static context and continuous content. Standard approach to image comparing based on pixels comparing only compares image content which makes it unusable for dynamic images. As part of this Master’s Thesis, an object detection algorithm was designed based on the separation and comparison of key points. The algorithm uses ORB algorithm for key points detection. The main problem that appears is false key point detection. False key point detection happens when algorithm compares referent key point with test key point and gets a positive result although it’s a wrong key point (mostly happens on images with text where same letters appear on multiple locations). That problem is solved by defining objects in a way that each key point is given an object (group) to which it belongs and then filtering all matches with wrong objects. Algorithm has been tested for distortion tolerance (rotation, scaling, particles and translation) and work efficiency on real actual sequences.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:388356
Study programmeTitle: Graduate University Study Programme in Electrical Engineering; specializations in: Communications and Informatics, Power Engineering
Course: Communications and Informatics
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike (magistar/magistra inženjer/inženjerka elektrotehnike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2016-11-29 08:59:42