prikaz prve stranice dokumenta Obrada videa računalnim vidom i umjetnom inteligencijom
Access restricted to authenticated users
undergraduate thesis
Obrada videa računalnim vidom i umjetnom inteligencijom
Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. urn:nbn:hr:179:645966
Svaguša, Luka
University of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
Department of Electronics and Computing

Institutional repository: Repository of the Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture

Cite this document

Svaguša, L. (2022). Obrada videa računalnim vidom i umjetnom inteligencijom (Undergraduate thesis). Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture. Retrieved from https://urn.nsk.hr/urn:nbn:hr:179:645966

Svaguša, Luka. "Obrada videa računalnim vidom i umjetnom inteligencijom." Undergraduate thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:179:645966

Svaguša, Luka. "Obrada videa računalnim vidom i umjetnom inteligencijom." Undergraduate thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, 2022. https://urn.nsk.hr/urn:nbn:hr:179:645966

Svaguša, L. (2022). 'Obrada videa računalnim vidom i umjetnom inteligencijom', Undergraduate thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, accessed 04 July 2024, https://urn.nsk.hr/urn:nbn:hr:179:645966

Svaguša L. Obrada videa računalnim vidom i umjetnom inteligencijom [Undergraduate thesis]. Split: University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture; 2022 [cited 2024 July 04] Available at: https://urn.nsk.hr/urn:nbn:hr:179:645966

L. Svaguša, "Obrada videa računalnim vidom i umjetnom inteligencijom", Undergraduate thesis, University of Split, Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture, Split, 2022. Available at: https://urn.nsk.hr/urn:nbn:hr:179:645966

Metadata
TitleObrada videa računalnim vidom i umjetnom inteligencijom
Title (english)Video processing using computer vision and artificial intelligence
AuthorLuka Svaguša
MentorSven Gotovac (mentor)
GranterUniversity of Split
Faculty of Electrical Engineering, Mechanical Engineering and Naval Architecture
(Department of Electronics and Computing)
Split
Defense date and country2022-09-16, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Electronics
Abstract
U sklopu ovog rada prikazana je i objašnjena realizacija nadzornog sustava za Trajektnu luku Split, ostvarenog implementacijom računalnog vida. Baza slika za treniranje modela na kojemu je sustav zasnovan stvorena je koristeći snimke nadzorne kamere sa vrha zgrade Koteks. Algoritam odabran za detekciju je YOLOv5s. Nakon treniranja modela ostvarene su sljedeće mjere uspješnosti : mAP_0.5 = 96.4%, mAP_0.5:0.95 = 75.1%. Na temelju analize preciznosti i odziva odabran je prag pouzdanosti od 60%. Za kraj je provedeno testiranje sustava na videozapisima s kojima se sustav prije nije susretao, te je ustanovljeno da model detektira prisutne brodove uz pretežno ispravne klasifikacije.
Abstract (english)
This thesis shows and explains the implementation of a surveillance system for the Split ferry port achieved through implementing computer vision. The image dataset used for training the model the system is based on was made using videos filmed by the surveillance camera from the top of the Koteks building. The algorithm chosen for detection was YOLOv5s. The following metrics were accomplished by training the model : mAP_0.5 = 96.4%, mAP_0.5:0.95 = 75.1%. Based on the analysis of the precision and recall, a confidence threshold of 60% was chosen. As the last part of the thesis, testing the trained model was carried out using yet unseen videos, it was concluded that the model detects present ships, and does it with the mostly correct classifications.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:179:645966
Study programmeTitle: Electrical Engineering and Information Technology
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka elektrotehnike i informacijske tehnologije)
Type of resourceText
File originBorn digital
Access conditionsAccess restricted to authenticated users
Terms of useLicence In copyright icon
Created on2023-07-05 12:04:11