Semantička segmentacija primjenom konvolucijskih neuronskih mreža

undergraduate thesis

• Title (english): Semantic Segmentation by Using Convolutional Neural Networks
• Author: Erik Matošević
• Mentor: Zoran Kalafatić (mentor)
• Committee member: Zoran Kalafatić (predsjednik povjerenstva)
• Committee member: Zvonimir Pavlić (član povjerenstva)
• Committee member: Ivan Filković (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-07-13, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Konvolucijske neuronske mreže su se pokazale kao jedno od najboljih rješenja u prepoznavanju objekata na slikama, i to ne samo u određivanju oznake za cjelokupnu sliku, već i u određivanju strukturiranih oznaka na izlazu za sliku dobivenu na ulazu. U ovome radu opisana je općenita arhitektura, te način rada konvolucijskih neuronskih mreža. Prikazan je problem semantičke segmentacije, te način na koji se konvolucijske neuronske mreže koriste u rješavanju istog. Uz rad je razvijena implementacija konvolucijske neuronske mreže koja je trenirana na skupu slika s nogometne utakmice, te su dobiveni rezultati opisani i prikazani.

Abstract (english)

Convolutional neural networks are at the forefront of recognizing objects in the images, and not only at whole-image classification, but also in determining structured output of tags for given input image. This work describes the general architecture of convolutional neural networks and the way they work. The problem of semantic segmentation and the way in which convolutional neural networks are used in solving the same was described. In addition to this work, the implementation of the convolutional neural network has been developed, which was trained in a collection of images from the football game, and the results are described and presented.

• Keywords: računalni vid; umjetne neuronske mreže; duboke neuronske mreže; konvolucijske neuronske mreže; semantička segmentacija; raspoznavanje objekata
• Keywords (english): computer vision; artificial neural networks; deep neural networks; convolutional neural networks; semantic segmentation; object recognition
• Language: croatian
• URN:NBN: urn:nbn:hr:168:006804
• 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: 2019-03-13 22:53:26