| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font
First page > Show document

Show document Help

Title:Prepoznava divjih živali na slikah z uporabo rezidualnih nevronskih mrež : diplomsko delo
Authors:ID Lakič, Mitja (Author)
ID Karakatič, Sašo (Mentor) More about this mentor... New window
Files:.pdf VS_Lakic_Mitja_2020.pdf (2,67 MB)
MD5: 4888B609D2D1BE3A65438EC19CF3547C
PID: 20.500.12556/dkum/bd0c8c77-44d6-4e18-af5c-5ae85e40e00b
 
Language:Slovenian
Work type:Bachelor thesis/paper
Typology:2.11 - Undergraduate Thesis
Organization:FERI - Faculty of Electrical Engineering and Computer Science
Abstract:V diplomskem delu se ukvarjamo s prepoznavanjem divjih živali na slikah z uporabo rezidualnih nevronskih mrež. Namen diplomskega dela je predstaviti rezidualne nevronske mreže in probleme, ki jih te mreže rešujejo. Pri prepoznavanju živali smo se omejili na 10 različnih kategorij divjih živali, podatkovna množica pa je bila sestavljena iz 10.000 slik. Rešitev smo razvili s pomočjo programskega jezika Python in programske knjižnice PyTorch. Primerjali smo rezultate treh različnih modelov nevronskih mrež, kjer je najboljši model dosegel 99,9-% točnost prepoznavanja. Ugotovili smo, da rezidualne nevronske mreže z uporabo preskočnih povezav zelo ugodno vplivajo na točnost modela, pri tem pa se najbolje izkažejo modeli, ki so bili predhodno naučeni.
Keywords:rezidualna nevronska mreža, računalniški vid, prepoznavanje divjih živali, globoko učenje, izginjajoči gradient
Place of publishing:Maribor
Place of performance:Maribor
Publisher:[M. Lakič]
Year of publishing:2020
Number of pages:IX, 65 f.
PID:20.500.12556/DKUM-76870 New window
UDC:004.932:004.8(043.2)
COBISS.SI-ID:38054915 New window
NUK URN:URN:SI:UM:DK:W8ONNTN1
Publication date in DKUM:03.11.2020
Views:888
Downloads:112
Metadata:XML DC-XML DC-RDF
Categories:KTFMB - FERI
:
LAKIČ, Mitja, 2020, Prepoznava divjih živali na slikah z uporabo rezidualnih nevronskih mrež : diplomsko delo [online]. Bachelor’s thesis. Maribor : M. Lakič. [Accessed 27 March 2025]. Retrieved from: https://dk.um.si/IzpisGradiva.php?lang=eng&id=76870
Copy citation
  
Average score:
0.5
1
1.5
2
2.5
3
3.5
4
4.5
5
(0 votes)
Your score:Voting is allowed only for logged in users.
Share:Bookmark and Share


Hover the mouse pointer over a document title to show the abstract or click on the title to get all document metadata.

Licences

License:CC BY-NC-ND 4.0, Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International
Link:http://creativecommons.org/licenses/by-nc-nd/4.0/
Description:The most restrictive Creative Commons license. This only allows people to download and share the work for no commercial gain and for no other purposes.
Licensing start date:21.07.2020

Secondary language

Language:English
Title:Recognizing wild animals on pictures with the use of residual neural networks
Abstract:In this thesis we are dealing with the recognition of wild animals on images using residual neural networks. The purpose of this thesis is to present residual neural networks and the problems that these networks solve. We have limited ourselves to 10 different categories of wild animals, the dataset consisted of 10.000 images. We developed our solution using the Python programming language and the PyTorch library. We compared the results of three different neural network models, the best model achieved a recognition accuracy of 99,9 %. We have noticed that residual neural networks, using skip connections, have a very positive effect on the accuracy of the model. Models that have been pretrained proved to yield the best results.
Keywords:residual neural network, computer vision, recognizing wild animals, deep learning, vanishing gradient


Comments

Leave comment

You must log in to leave a comment.

Comments (0)
0 - 0 / 0
 
There are no comments!

Back
Logos of partners University of Maribor University of Ljubljana University of Primorska University of Nova Gorica