Sustav za detekciju vozila u video snimkama prometnica

undergraduate thesis

• Title (english): System for detecting vehicles in traffic video sequences
• Author: Leonard Pačandi
• Mentor: Zoran Kalafatić (mentor)
• Committee member: Zoran Kalafatić (predsjednik povjerenstva)
• Committee member: Hrvoje Nuić (član povjerenstva)
• Committee member: Zorana Doždor (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Detekcija objekata je jedan od temeljnih problema kojim se bavi računalni vid. Sve više se počinje implementirati u sve aspekte našeg života, naročito automonmu vožnju i inteligentne sustave za praćenje i kontrolu prometa. U ovom radu je opisan jedan način treniranja modela, koristeći YOLOv5 algoritam. S obzirom da je YOLOv5 predtreniran, ne trebamo jako veliki skup podataka u slučaju da želimo da sustav detektira neki relativno običan objekt, u našem slučaju vozilo. Prikazani su rezultati modela koji je treniran na skupu podataka grada Montreala. Ubuduće, za bolje rezultate bi svakako trebalo koristiti veći skup podataka.

Abstract (english)

Object detection is one of the fundamental problems of computer vision. It is constantly being implemented in all aspects for our lives, particularly autonomous vehicles and intelligent systems for tracking and controlling traffic. In this paper, one way of training the model is described, using the YOLOv5 algorithm. Considering the fact that YOLOv5 is pretrained, we do not need a very big dataset to train a relatively simple object, which in our case is a vehicle. The results presented are of a model trained on a dataset of the city of Montreal. In the future, for better results a larger dataset should be used.

• Keywords: detekcija vozila; konvolucijske neuronske mreže; YOLOv5 algoritam; inteligentno praćenje prometa
• Keywords (english): vehicle detection; convolutional neural networks; YOLOv5 algorithm; intelligent traffic tracking
• Language: croatian
• URN:NBN: urn:nbn:hr:168:253192
• 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: Closed access
• Created on: 2023-07-13 10:25:23