Razvoj aplikacije za detekciju slobodnih mjesta na parkiralištu korištenjem dubokog učenja

undergraduate thesis

• Title (english): Development of deep learning-based application for detecting vacant parking spaces
• Author: Luka Babić
• Mentor: Marko Horvat (mentor)
• Committee member: Marko Horvat (predsjednik povjerenstva)
• Committee member: Juraj Dončević (član povjerenstva)
• Committee member: Kristina Ćaleta (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2024-07-04, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Detekcija vozila postupak je računalnog vida pomoću korištenja alata poput dubokih
neuronskih mreža. Postoje različite mreže koje je moguće primijeniti na ovakav ili sličan
problem, a u ovom radu korištena je duboka neuronska mreža YOLOv8.
U implementaciji ovog rješenja, korišteno je radno okruženje Google Colab, te Python
programski jezik uz biblioteku ultralytics i upervision.
Aplikacija omogućava učitavanje fotografija i videozapisa s Google Drive-a, procjenu
položaja vozila na... More fotografijama, te spremanje rezultata nazad na Google Drive.
Analiza je provedena pomoću standardne konfuzijske matrice, te pratećih mjera kvalitete
performansi: točnost (engl. accuracy), odziv (engl. recall), preciznost (engl. precision) i
F1-score.
Less

Abstract (english)

Vehicle detection is a computer vision process using tools like deep neural networks. There
are different networks that can be applied to this or a similar problem, and in this work the
YOLOv8 deep neural network was used.
In the implementation of this solution, the Google Colab working environment and the
Python programming language with the Ultralytics and Supervision library were used.
The application allows uploading photos and videos from Google Drive, estimating... More the
position of the car in the photos, and saving the results back to Google Drive.
The analysis was carried out using the standard confusion matrix, and accompanying
performance quality measures: accuracy, recall, precision and F1-score. Less

• Keywords: konvolucijske neuronske mreže; duboko učenje; računalni vid; detekcija parking mjesta; YOLOv8
• Keywords (english): convolutional neural networks; deep learning; computer vision; parking space detection; YOLOv8
• Language: croatian
• URN:NBN: urn:nbn:hr:168:612553
• Study programme: Title: Computing; Study programme type: university; Study level: undergraduate; Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Open access
• Created on: 2024-10-12 20:00:11