Back to search

Institutional Repository of Vilnius University

Neuroninio tinklo atsparumas transformacijoms /

Abstract

dc:description

Neurol Network Resistence to Tranformations One of the widely studied problems of computer vision is the discovery of similarities and differences between the two images. Neural networks, which is trained by large amounts of data, are also used in this area. Neural networks can solve this problem faster compared to other solutions. Unmanned aerial vehicles (drones) face the problem of positioning in places inaccessible to satellite signals. To solve this problem, cameras in drones can be used to take photos of which are compared with a topographic map using a trained neural network. The photos taken may not match the topographic map as they will be distorted by the camera lens and the shooting angle. Therefore, in this thesis, possible image distortions are simulated and a study of the effects of distortions on a trained neural network model is performed.

Degree

thesis:*
Grantor dc:publisher
Institutional Repository of Vilnius University
Year dc:date
2020

Author and committee

dc:creator, dc:contributor.*
Author dc:creator
  • Povilonis, Lukas,
Contributors dc:contributor
  • Valaitis, Vytautas

Rights

dc:rights
Statement dc:rights
  • info:eu-repo/semantics/openAccess
Language dc:language
lit

Identifiers

dc:identifier.*

Chain of custody

source
Harvested from
Vilnius University
Base URL
epublications.vu.lt/oai
Last updated
2026-11-19
Source record
OAI-PMH GetRecord
related terms
citation

Povilonis, Lukas,. Neuroninio tinklo atsparumas transformacijoms /. Institutional Repository of Vilnius University, 2020. https://repository.vu.lt/VU:ELABAETD81705233&prefLang=en_US