prikaz prve stranice dokumenta Uspoređivanje dinamičnih slika na osnovi karakteristika objekta
  Download
PDF 2.53 MB
master's thesis
Uspoređivanje dinamičnih slika na osnovi karakteristika objekta
2024. urn:nbn:hr:200:240444
Lučić, Luka
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Communications
Chair of Multimedia Systems and Digital Television

Cite this document

Lučić, L. (2024). Uspoređivanje dinamičnih slika na osnovi karakteristika objekta (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:200:240444

Lučić, Luka. "Uspoređivanje dinamičnih slika na osnovi karakteristika objekta." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. https://urn.nsk.hr/urn:nbn:hr:200:240444

Lučić, Luka. "Uspoređivanje dinamičnih slika na osnovi karakteristika objekta." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2024. https://urn.nsk.hr/urn:nbn:hr:200:240444

Lučić, L. (2024). 'Uspoređivanje dinamičnih slika na osnovi karakteristika objekta', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 28 October 2024, https://urn.nsk.hr/urn:nbn:hr:200:240444

Lučić L. Uspoređivanje dinamičnih slika na osnovi karakteristika objekta [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2024 [cited 2024 October 28] Available at: https://urn.nsk.hr/urn:nbn:hr:200:240444

L. Lučić, "Uspoređivanje dinamičnih slika na osnovi karakteristika objekta", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2024. Available at: https://urn.nsk.hr/urn:nbn:hr:200:240444

Please login to the repository to save this object to your list.
Metadata
TitleUspoređivanje dinamičnih slika na osnovi karakteristika objekta
Title (english)Comparison of dynamic images based on object characteristics
AuthorLuka Lučić
MentorMario Vranješ (mentor)
Committee memberMario Vranješ (predsjednik povjerenstva)
Committee memberDenis Vranješ (član povjerenstva)
Committee memberRatko Grbić (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Communications)
(Chair of Multimedia Systems and Digital Television)
Osijek
Defense date and country2024-10-01, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Data Processing
Abstract
U ovom diplomskom radu razvijen je i testiran unaprijeđeni algoritam za usporedbu dinamičkih slika, zasnovan na DISM algoritmu. Glavni cilj bio je poboljšati točnost prepoznavanja sličnosti slika proširivanjem metodologije koja obuhvaća detekciju ključnih točaka, analizu kontura i boje objekata. Verifikacija algoritma provedena je na CGIAD i CG-STB-DI bazama slika, a rezultati su pokazali da unaprijeđeni algoritam zadržava visoku razinu točnosti originalnog DISM algoritma i pruža dodatnu preciznost u složenim scenarijima. Točnost unaprijeđenog algoritma je 100% za CGIAD i CG-STB-DI baze slika, a vrijeme obrade na CGIAD bazi slika se, u odnosu na originalni DISM algoritam, povećalo za 2,23%. Ključna razlika između algoritma zasnovanog na sijamskoj neuronskoj mreži, originalnog i unaprijeđenog algoritma leži u sveobuhvatnijoj analizi slika koja uključuje više faktora, što rezultira boljim performansama u različitim uvjetima. Najmanje vrijeme, ali zato i najmanju točnost, postigao je algoritam zasnovan na sijamskoj neuronskoj mreži. S obzirom na originalni algoritam, unaprijeđeni algoritam daje veću točnost, ali i nešto veći utrošak vremena za obradu CGIAD baze slika. Ovaj rad predstavlja značajan doprinos području računalnog vida, postavljajući temelje za daljnji razvoj i primjenu algoritama za usporedbu dinamičnih slika u raznim industrijama.
Abstract (english)
In this thesis, an improved algorithm for comparing dynamic images, based on the DISM algorithm, was developed and tested. The main goal was to improve the accuracy of image similarity recognition by extending the methodology to include keypoint detection, contour analysis, and object color. Algorithm verification was performed on CGIAD and CG-STB-DI image databases, and the results showed that the improved algorithm maintains the high level of accuracy of the original DISM algorithm and provides additional precision in complex scenarios. The accuracy of the improved algorithm is 100% for the CGIAD and CG-STB-DI image bases, and the processing time on the CGIAD image base, compared to the original DISM algorithm, has increased by 2.23%. The key difference between the Siamese neural network-based algorithm, the original algorithm, and the improved algorithm lies in the more comprehensive image analysis that includes multiple factors, resulting in better performance under different conditions. The least time, but therefore the least accuracy, was achieved by an algorithm based on a Siamese neural network. With regard to the original algorithm, the improved algorithm provides higher accuracy, but also a slightly higher time consumption for processing the CGIAD image database. This work represents a significant contribution to the field of computer vision, laying the foundations for further development and application of dynamic image comparison algorithms in various industries.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:240444
Study programmeTitle: Graduate university study programme in Electrical Engineering (branches: Power Engineering and Communications and Informatics; specializations in: Engineering, Power Engineering and Communications, Communications and Informatics
Course: Communications and Informatics
Study programme type: university
Study level: graduate
Academic / professional title: sveučilišni/a magistar/magistra inženjer/inženjerka elektrotehnike (sveučilišni/a magistar/magistra inženjer/inženjerka elektrotehnike)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2024-10-04 10:59:56