diplomsko delo
Abstract
Diplomsko delo predstavlja implementacijo trenutno vodilne metode za geolokalizacijo brezpilotnih letalnikov, ob izgubi sistema za določanje položaja; implementacija ni bila javno dostopna. V okviru dela smo ustvarili novo podatkovno zbirko, ki vsebuje pare slik iz brezpilotnega letalnika in pripadajočih satelitskih posnetkov. Osredotočili smo se na uporabo naprednih nevronskih mrež, zlasti na piramidni vision transformer (PVT). Ključno vlogo je imela siamska nevronska mreža za primerjavo vzorcev med obema vrstama slik. Metodologija je bila podprta z različnimi optimizacijskimi strategijami, vključno z uporabo stratificiranega vzorčenja, Hanningovega okna in regularizacijskih tehnik. Rezultati potrjujejo učinkovitost predlagane metode za natančno geolokalizacijo brezpilotnih letalnikov. Delo zaključujemo s poudarkom na ključnih ugotovitvah in potencialu razvite metode.
Keywords
lokalizacija brezpilotnih letalnikov;geo-lokalizacija;transformer;računalništvo in informatika;univerzitetni študij;diplomske naloge;
Data
Language: |
Slovenian |
Year of publishing: |
2023 |
Typology: |
2.11 - Undergraduate Thesis |
Organization: |
UL FRI - Faculty of Computer and Information Science |
Publisher: |
[G. Spagnolo] |
UDC: |
004.8:629.014.9(043.2) |
COBISS: |
168320771
|
Views: |
135 |
Downloads: |
31 |
Average score: |
0 (0 votes) |
Metadata: |
|
Other data
Secondary language: |
English |
Secondary title: |
Unmanned aerial vehicle localization using satellite images |
Secondary abstract: |
The thesis presents the implementation of the currently leading method for
geolocation of unmanned aerial vehicles, in the event of a loss of the positioning system; the implementation was not publicly available. As part of the work, we created a new dataset containing pairs of images from unmanned aerial vehicles and corresponding satellite images. We focused on the use of advanced neural networks, especially the pyramid vision transformer (PVT). A key role was played by the siamese neural network for comparing patterns between the two types of images. The methodology was supported by various
optimization strategies, including the use of stratified sampling, the Hanning window, and regularization techniques. The results confirm the effectiveness of the proposed method for accurate geolocation of unmanned aerial vehicles. We conclude the work with an emphasis on key findings and the potential of the developed method. |
Secondary keywords: |
unmanned aerial vehicle localization;geo-localization;deep learning;transformer;computer science;computer and information science;diploma;Brezpilotni letalniki;Teledetekcijski posnetki;Globoko učenje (strojno učenje);Računalništvo;Univerzitetna in visokošolska dela; |
Type (COBISS): |
Bachelor thesis/paper |
Study programme: |
1000468 |
Embargo end date (OpenAIRE): |
1970-01-01 |
Thesis comment: |
Univ. v Ljubljani, Fak. za računalništvo in informatiko |
Pages: |
64 str. |
ID: |
19933619 |