prikaz prve stranice dokumenta Semantičko mapiranje za manipuliranje u interakciji s okolinom
No public access
master's thesis
Semantičko mapiranje za manipuliranje u interakciji s okolinom
Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. urn:nbn:hr:168:872468
Kunštek, Matija
University of Zagreb
Faculty of Electrical Engineering and Computing

Institutional repository: FER Repository

Cite this document

Kunštek, M. (2020). Semantičko mapiranje za manipuliranje u interakciji s okolinom (Master's thesis). Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing. Retrieved from https://urn.nsk.hr/urn:nbn:hr:168:872468

Kunštek, Matija. "Semantičko mapiranje za manipuliranje u interakciji s okolinom." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. https://urn.nsk.hr/urn:nbn:hr:168:872468

Kunštek, Matija. "Semantičko mapiranje za manipuliranje u interakciji s okolinom." Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, 2020. https://urn.nsk.hr/urn:nbn:hr:168:872468

Kunštek, M. (2020). 'Semantičko mapiranje za manipuliranje u interakciji s okolinom', Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, accessed 22 November 2024, https://urn.nsk.hr/urn:nbn:hr:168:872468

Kunštek M. Semantičko mapiranje za manipuliranje u interakciji s okolinom [Master's thesis]. Zagreb: University of Zagreb, Faculty of Electrical Engineering and Computing; 2020 [cited 2024 November 22] Available at: https://urn.nsk.hr/urn:nbn:hr:168:872468

M. Kunštek, "Semantičko mapiranje za manipuliranje u interakciji s okolinom", Master's thesis, University of Zagreb, Faculty of Electrical Engineering and Computing, Zagreb, 2020. Available at: https://urn.nsk.hr/urn:nbn:hr:168:872468

Metadata
TitleSemantičko mapiranje za manipuliranje u interakciji s okolinom
Title (english)Semantic mapping for manipulation through interaction
AuthorMatija Kunštek
MentorMatko Orsag (mentor)
Committee memberMatko Orsag (predsjednik povjerenstva)
Committee memberStjepan Bogdan (član povjerenstva)
Committee memberZdenko Kovačić (član povjerenstva)
GranterUniversity of Zagreb
Faculty of Electrical Engineering and Computing
Zagreb
Defense date and country2020-09-21, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Electrical Engineering
Abstract
U ovom diplomskom radu istražena je i implementirana mreža dubokog učenja na način da se može upotrijebiti u zadatku semantičke segmentacije. Svrha toga je povećanje autonomije u području agrikulture. Kako bi se korištena mreža adaptirala i evaluirala definirali su se odgovarajući skupovi podataka s različitim evaluacijskim zadacima. Za implementaciju odabrane neuronske mreže koristio se Python programski jezik u kombinaciji s TensorFlow platformom, dok su Capsicum annuum i fikus biljke korištene u obliku skupa podataka za treniranje i testiranje. Kako bi se testirale mogućnosti povezivanja mreže s korištenim robotskim manipulatorom primijenilo se Gazebo simulacijsko okruženje u kombinaciji s ROS operacijskim sustavom. U konačnici su i predložene mogućnosti za daljnje unapređivanje autonomije u području agrikulture.
Abstract (english)
In this master's thesis, a deep learning neural network is explored and implemented in such a way that it can be used in a semantic image segmentation task to increase autonomy in the field of agriculture. In order to adapt and evaluate the utilized network in the field of agriculture, appropriate datasets with various evaluation tasks were defined. For the implementation of the selected deep learning neural network, the Python programming language combined with the TensorFlow machine learning platform was utilized, while the Capsicum annuum and ficus plants were employed in the form of training and testing datasets. In order to assess the network's feasibility with the used robotic manipulator, the Gazebo simulation environment with the ROS operating system was utilized. Finally, possibilities for further improvements in agriculture were proposed.
Keywords
Keywords (english)
Languageenglish
URN:NBNurn:nbn:hr:168:872468
Study programmeTitle: Electrical Engineering and Information Technology
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije (magistar/magistra inženjer/inženjerka elektrotehnike i informacijske tehnologije)
Type of resourceText
File originBorn digital
Access conditionsClosed access
Terms of useLicence In copyright icon
Public note
Created on2023-01-13 12:50:57