prikaz prve stranice dokumenta Algoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima
  Download
PDF 3.07 MB
master's thesis
Algoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima
Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. urn:nbn:hr:200:664056
Čordašić, Ivan
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

Čordašić, I. (2023). Algoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima (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:200:664056

Čordašić, Ivan. "Algoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:664056

Čordašić, Ivan. "Algoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:664056

Čordašić, I. (2023). 'Algoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 15 March 2025, https://urn.nsk.hr/urn:nbn:hr:200:664056

Čordašić I. Algoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2023 [cited 2025 March 15] Available at: https://urn.nsk.hr/urn:nbn:hr:200:664056

I. Čordašić, "Algoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:200:664056

Please login to the repository to save this object to your list.
Metadata
TitleAlgoritam za izdvajanje putokaza iz scene i razumijevanje teksta na njima
Title (english)Algorithm for extracting signposts from the scene and understanding the text on them
AuthorIvan Čordašić
MentorMario Vranješ (mentor)
Committee memberMario Vranješ (predsjednik povjerenstva)
Committee memberMarijan Herceg (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 country2023-09-26, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Telecommunications and Informatics
Abstract
U sklopu ovog rada razvijen je algoritam za detekciju putokaza u sceni i razumijevanje teksta na njima, koristeći vlastitu bazu podataka. Cilj je bio izolirati putokaze iz scene i razumjeti tekst na njima, uključujući smjer na koji pokazuju. Algoritam je izgrađen koristeći metode dubokog učenja, odnosno YOLOv7 za detekciju putokaza i smjera na njima, u kombinaciji s OCR alatima za razumijevanje teksta. Razvoj algoritma proveden je u Python programskom okruženju. Testiranja su provedena na... More različitim slikama, koje su snimljene u različitim uvjetima. Rezultati su pokazali visoku točnost detekcije putokaza i smjera na njime te razumijevanja teksta, ali i prostor za poboljšanje u situacijama kada su putokazi djelomično zaklonjeni ili kada je kvaliteta slike niska. Algoritam predstavlja dobar korak prema poboljšanju navigacijskih sustava kroz detekciju i interpretaciju putokaza. Iako su rezultati obećavajući, daljnja optimizacija je potrebna kako bi se poboljšala performansa u realnom vremenu i u različitim uvjetima na cesti. U okviru testiranja, posebno je važno istaknuti primjenu implementiranih algoritama na Raspberry Pi 4 ugradbenoj računalnoj platformi. Iako su početni rezultati obećavajući, primjetna je potreba za dodatnom optimizacijom kako bi se osigurala pouzdana i brza obrada slika u stvarnom vremenu zbog softverskih i hardverskih ograničenja ugradbene računalne platforme. To uključuje optimizaciju hardverskih resursa, kao i prilagodbu algoritama kako bi se što efikasnije koristili ograničeni resursi ugradbenih računalnih platformi. Činjenica koja ide u prilog tom zaključku dolazi iz usporedbe performansi rada na osobnom računalu i ugradbenoj računalnoj platformi, gdje osobno računalo pokazuje bolje performanse te je veći spektar algoritama koji se mogu koristiti. Less
Abstract (english)
As part of this work, an algorithm was developed for the detection of road signs in the scene and the understanding of the text on them, using our own database. The goal was to isolate the signposts from the scene and understand the text on them, including the direction they point to. The algorithm was built using deep learning methods, namely YOLOv7 to detect road signs and directions on them, combined with OCR tools for text understanding. The development of the algorithm was carried out... More in the Python programming environment. The tests were carried out on different photos, which were taken in different conditions. The results showed a high accuracy of the detection of road signs and the direction on them, as well as understanding of the text, but also room for improvement in situations where the road signs are partially obscured or when the image quality is low. The algorithm represents a good step towards the improvement of navigation systems through the detection and interpretation of road signs. Although the results are promising, further optimization is needed to improve real-time performance under different road conditions. As part of the testing, it is particularly important to highlight the application of the implemented algorithms on the Raspberry Pi 4 embedded computer platform. Although the initial results are promising, there is a noticeable need for further optimization to ensure reliable and fast real-time image processing due to the software and hardware limitations of the embedded computer platform. This includes the optimization of hardware resources, as well as the adaptation of algorithms to make the most efficient use of the limited resources of embedded computer platforms. The fact that supports this conclusion comes from a comparison of the performance of work on a personal computer and on an embedded computer platform, where the personal computer shows better performance and the spectrum of algorithms that can be used is greater. Less
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:664056
Study programmeTitle: UNIVERSITY STUDY AUTOMOBILE COMPUTER AND COMMUNICATION
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka automobilskog računarstva i komunikacija (magistar/magistra inženjer/inženjerka automobilskog računarstva i komunikacija)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
RepositoryFaculty of Electrical Engineering, Computer Science and Information Technology Osijek
Created on2023-10-11 08:54:16
accessibility

closeAccessibilityrefresh

If you wish to permanently save changes, click on Save, if not - your settings will be reset when you restart the browser.