Klasifikacija krompirja s pomočjo globokih nevronskih mrež s podporo barvne in termovizijske analize : magistrsko delo

Pec, Taja

| | SLO | ENG | Cookies and privacy

Bigger font | Smaller font

First page > Show document

Show document

Title:	Klasifikacija krompirja s pomočjo globokih nevronskih mrež s podporo barvne in termovizijske analize : magistrsko delo
Authors:	ID Pec, Taja (Author) ID Šafarič, Riko (Mentor) More about this mentor... ID Klančnik, Simon (Mentor) More about this mentor...
Files:	MAG_Pec_Taja_2025.pdf (8,18 MB) MD5: 871950C7493D2D419D4D6DAD7B119530
Language:	Slovenian
Work type:	Master's thesis/paper
Typology:	2.09 - Master's Thesis
Organization:	FERI - Faculty of Electrical Engineering and Computer Science
Abstract:	V magistrskem delu obravnavamo klasifikacijo krompirja v tri kakovostne razrede – gnili, krmni in jedilni – z uporabo naprednih globokih nevronskih mrež. Model smo razvili v programskem jeziku Python z uporabo ogrodja TensorFlow. Primerjali smo učinkovitost treh sodobnih arhitektur konvolucijskih nevronskih mrež: EfficientNet, DenseNet in Xception, ter na podlagi rezultatov izbrali najbolj primerno za našo podatkovno bazo. DenseNet201 je izstopal kot najbolj natančen in stabilen model, DenseNet121 pa je ponujal najboljše ravnovesje med natančnostjo in računsko zahtevnostjo. Preizkusili smo tudi termovizijsko kamero za preučevanje možnosti zaznave gnilobe na podlagi temperaturnih razlik krompirja in opisali omejitve te metode. Cilj raziskave je razvoj inteligentnega, cenovno dostopnega sistema za avtomatsko sortiranje krompirja, ki bi povečal produktivnost kmetijske proizvodnje ob minimalnih stroških.
Keywords:	globoko učenje, konvolucijske nevronske mreže, klasifikacija krompirja, TensorFlow, termovizijska analiza
Place of publishing:	Maribor
Place of performance:	Maribor
Publisher:	[T. Pec]
Year of publishing:	2025
Number of pages:	1 spletni vir (1 datoteka PDF (XI, 93 str.))
PID:	20.500.12556/DKUM-94623
UDC:	004.383.8:633.491(043.2)
COBISS.SI-ID:	260198659
Publication date in DKUM:	22.09.2025
Views:	145
Downloads:	35
Metadata:
Categories:	KTFMB - FERI
:	Copy citation

Average score:	(0 votes)
Your score:	Voting is allowed only for logged in users.
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:	22.08.2025

Secondary language

Language:	English
Title:	Classification of potatoes using deep neural networks with support of color and thermal imaging analysis
Abstract:	This master's thesis addresses the classification of potatoes into three quality categories – rotten, feed and edible – using advanced deep neural networks. The model was developed in the Python programming language utilizing the TensorFlow framework. We compared the performance of three modern convolutional neural network architectures: EfficientNet, DenseNet and Xception, and selected the most suitable one for our dataset based on the results. DenseNet201 stood out as the most accurate and stable model, while DenseNet121 offered the best balance between accuracy and computational complexity. We also tested a thermal imaging camera to explore the possibility of detecting rot based on temperature differences in potatoes and described the limitations of this method. The aim of the research is to develop an intelligent, cost-effective system for automatic potato sorting that would increase agricultural productivity with minimal costs.
Keywords:	deep learning, convolutional neural networks, potato classification, TensorFlow, thermal imaging analysis

Comments

Leave comment

You must log in to leave a comment.

Comments (0)

0 - 0 / 0

There are no comments!

Back