prikaz prve stranice dokumenta Primjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu
Access restricted to students and staff of home institution
undergraduate thesis
Primjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu
2022. urn:nbn:hr:166:259651
Čokić, Valentina
University of Split
Faculty of Science
Department of Informatics

Cite this document

Čokić, V. (2022). Primjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu (Undergraduate thesis). Split: University of Split, Faculty of Science. Retrieved from https://urn.nsk.hr/urn:nbn:hr:166:259651

Čokić, Valentina. "Primjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu." Undergraduate thesis, University of Split, Faculty of Science, 2022. https://urn.nsk.hr/urn:nbn:hr:166:259651

Čokić, Valentina. "Primjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu." Undergraduate thesis, University of Split, Faculty of Science, 2022. https://urn.nsk.hr/urn:nbn:hr:166:259651

Čokić, V. (2022). 'Primjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu', Undergraduate thesis, University of Split, Faculty of Science, accessed 23 April 2024, https://urn.nsk.hr/urn:nbn:hr:166:259651

Čokić V. Primjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu [Undergraduate thesis]. Split: University of Split, Faculty of Science; 2022 [cited 2024 April 23] Available at: https://urn.nsk.hr/urn:nbn:hr:166:259651

V. Čokić, "Primjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu", Undergraduate thesis, University of Split, Faculty of Science, Split, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:166:259651

Please login to the repository to save this object to your list.
Metadata
TitlePrimjena YOLOv5 algoritma na prilagođenom skupu podataka u stvarnom vremenu
Title (english)Application of the YOLOv5 algorithm on a custom real time data set
AuthorValentina Čokić
MentorSaša Mladenović (mentor)
Committee memberSaša Mladenović (predsjednik povjerenstva)
Committee memberDivna Krpan (član povjerenstva)
Committee memberGoran Zaharija (član povjerenstva)
GranterUniversity of Split
Faculty of Science
(Department of Informatics)
Split
Defense date and country2022-09-26, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Abstract
Cilj je završnoga rada prikazati treniranje YOLOv5 algoritam na prilagođenome setu podataka kako bi algoritam mogao prepoznati slova hrvatske znakovne abecede. Na početku rada objašnjeno je što je detekcija objekata, kojom se rad bavi. U drugom dijelu rada objašnjena je konvolucijska neuronska mreža koja se nalazi u pozadini YOLO algoritma. Treći dio rada bavi se YOLO algoritmom. Opisana je kratka povijest nastanka tog algoritma i način na koji radi te je opisana arhitektura YOLOv5 algoritma. Četvrti dio rada objašnjava sve korake potrebne za implementaciju ovoga algoritma. Najprije je opisano prikupljanje i označavanje svih slika za skup podataka, zatim treniranje samoga modela te u konačnici instalacija algoritma na operativni sustav Windows kako bi se detekcija mogla izvoditi u stvarnome vremenu. Zadnji dio rada analizira dobivene rezultate te se predlažu određene smjernice za daljnji rad na projektu i za poboljšanje rezultata projekta.
Abstract (english)
The goal is to show the training of the YOLOv5 algorithm on a customized dataset so that the algorithm can recognize the letters of the Croatian sign alphabet. In the first part, it was explained what object detection is. In the second part, the convolutional neural network, which is behind the YOLO algorithm, is clarified. The third part deals with the YOLO algorithm. A history of the creation, the way it works and the architecture is explained. The fourth part explains all the steps necessary to implement this algorithm. First, the collection and labelling of all images are described, then the training, and finally, the installation of the algorithm on the Windows operating system so that the detection can be performed in real-time. The last part of the paper analyzes the obtained results and suggests specific guidelines for further work on the project and how to improve the results.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:166:259651
Study programmeTitle: Computer Science
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) informatike (sveučilišni/a prvostupnik/prvostupnica (baccalaureus/baccalaurea) informatike)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to students and staff of home institution
Terms of useLicence In copyright icon
Created on2023-02-27 10:28:40