Institutional Repository of Vilnius University
Ortografinių nuotraukų lyginimas naudojant gilųjį neuroninį tinklą /
Abstract
dc:descriptionUnmanned aerial vehicle automated control systems rely on GPS signals, but at times this signal can be lost due to environment. To solve this issue the technology of deep neural networks is employed. Using this technology, drones can locate themselves from aerial images. Nevertheless optics can distort the input images. The goal of this paper is to check if images with "fish eye" lens distortion can be recognized by a neural network that has been trained on undistorted images. 5 data sets were created with "fish eye" lens distortion. 2 models were prepared: one based on VGGNet architecture and another based on Inception-v3. Both models achieved 100\% accuracy with every dataset. But the VGGNet based model showed higher probabilities of recognition then Inception-v3 based model.
Degree
thesis:*- Grantor dc:publisher
- Institutional Repository of Vilnius University
- Year dc:date
- 2021
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Paulikas, Dominykas,
- Contributors dc:contributor
-
- Valaitis, Vytautas
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:ELABAETD107115837&prefLang=en_US