Semantička segmentacija prometnih scena

master's thesis

• Title (english): Semantic segmentation of traffic scenes
• Author: Antonio Vencl
• Mentor: Tomislav Hrkać (mentor)
• Committee member: Tomislav Hrkać (predsjednik povjerenstva)
• Committee member: Zoran Kalafatić (član povjerenstva)
• Committee member: Leonardo Jelenković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2023-07-10, Croatia

Abstract

U ovom radu dan je uvod u područje računalnog vida, napravljen je kratki uvod u biološku i umjetnu neuronsku mrežu. Ukratko su objašnjene konvolucijske neuronske mreže koje su česti alat u problemima računalnog vida, kao i proces učenja neuronskih mreža općenito Predstavljen je zadatak, te jedna implementacija koja rješava sličan zadatak. Predstavljeni su podaci koji su korišteni za treniranje i testiranje neuronske mreže. Objašnjen je korišteni model i koje su njegove karakteristike bitne za zadani problem. Te na kraju su pokazani rezultati na podacima za validiranje.

Abstract (english)

In this paper, an introduction to the field of computer vision was given, a short introduction to biological and artificial neural networks was made. Convolutional neural networks, which are a common tool in computer vision problems, are briefly explained, as well as the learning process of neural networks in general. The task is presented, as well as one implementation that solves a similar task. The data used for training and testing the neural network are presented. The used model is explained and which of its characteristics are important for the given problem. And at the end, the results on the validation data were shown.

• Keywords: neruonska mreža; konvolucija; vozila; pješaci; prometni znakovi; računalni vid; python
• Keywords (english): neural network; convolution; vehicles; pedestrians; road signs; computer vision; python
• Language: croatian
• URN:NBN: urn:nbn:hr:168:518451
• Study programme: Title: Computing; Study programme type: university; Study level: graduate; Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
• Type of resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2023-10-11 10:53:10