Institutional Repository of Vilnius University
Ortografinių multispektrinių nuotraukų detalus lyginimas /
Abstract
dc:descriptionDrones utilise GPS to determine their location, however this may not always be available. An alternative method to determine the location is to compare the photos produced by the drone with a map until the most relevant location is determined. Neural networks can be used to compare photos. In this study, neural models are trained using the VGG-16, MobileNet, ResNet, EfficientNet, EfficientNetV2 architectures used in previous research and new Res2Net and DLA architectures used as backbones by CenterNet2. The models are trained in the Austrian, Iraqi and Russian regions, and their results are analyzed.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2024
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Senda, Edvinas,
Rights
dc:rights- Statement dc:rights
-
- info:eu-repo/semantics/openAccess
- Language dc:language
- lit
Identifiers
dc:identifier.*- Repository record dc:identifier
- https://repository.vu.lt/VU:ELABAETD210647940&prefLang=en_US