prikaz prve stranice dokumenta Automatska detekcija voća
  Download
PDF 2.05 MB
master's thesis
Automatska detekcija voća
Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. urn:nbn:hr:200:612252
Perić, Luka
Josip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
Department of Computer Engineering and Automation
Chair of Automation and Robotics

Institutional repository: Faculty of Electrical Engineering, Computer Science and Information Technology Osijek

Cite this document

Perić, L. (2023). Automatska detekcija voća (Master's thesis). Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek. Retrieved from https://urn.nsk.hr/urn:nbn:hr:200:612252

Perić, Luka. "Automatska detekcija voća." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:612252

Perić, Luka. "Automatska detekcija voća." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2023. https://urn.nsk.hr/urn:nbn:hr:200:612252

Perić, L. (2023). 'Automatska detekcija voća', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 27 November 2024, https://urn.nsk.hr/urn:nbn:hr:200:612252

Perić L. Automatska detekcija voća [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2023 [cited 2024 November 27] Available at: https://urn.nsk.hr/urn:nbn:hr:200:612252

L. Perić, "Automatska detekcija voća", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2023. Available at: https://urn.nsk.hr/urn:nbn:hr:200:612252

Metadata
TitleAutomatska detekcija voća
Title (english)Automatic fruit detection
AuthorLuka Perić
MentorEmmanuel Karlo Nyarko (mentor)
Committee memberEmmanuel Karlo Nyarko (član povjerenstva)
Committee memberIvan Vidović (predsjednik povjerenstva)
Committee memberRobert Cupec (član povjerenstva)
GranterJosip Juraj Strossmayer University of Osijek
Faculty of Electrical Engineering, Computer Science and Information Technology Osijek
(Department of Computer Engineering and Automation)
(Chair of Automation and Robotics)
Osijek
Defense date and country2023-09-29, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Abstract
U ovom diplomskom radu testirano je nekoliko algoritama računalnog vida i umjetnih neuronskih mreža koje bi se mogle primijeniti pri prepoznavanju voća u kartonskim kutijama. Prikupljene su slike za podatkovni skup za prepoznavanje jabuka sorte Granny Smith. Podatkovni skup je anotiran i augmentiran. Za detektiranje jabuka korištene su Faster RCNN i YOLO arhitekture neuronskih mreža. Sve mreže su imale odlične evaluacijske metrike. Pri procjeni veličine jabuka korištena je segmentacija slike po boji. Segmentiranjem slike po boji, uz korištenje OpenCV biblioteke, prepoznaju se jabuke manjih dimenzija. Za prebrojavanje jabuka unutar kutije testirana je Houghova transformacija za krugove. Zbrojiti prepoznate jabuke (izlaze) koje daju umjetne neuronske mreže pokazao se kao bolji izbor za prebrojavanje jabuka unutar kutije. Testirani algoritmi mogu poslužiti pri kontroli kvalitete voća unutar kutija u industrijskim postrojenjima.
Abstract (english)
In this paper, several computer vision algorithms and artificial neural networks that could be applied in the recognition of fruit in cardboard boxes were tested. Images were collected for the Granny Smith apple recognition dataset. The dataset is annotated and augmented. Faster RCNN and YOLO neural network architectures were used to detect apples. All networks had excellent evaluation metrics. Color image segmentation was used to estimate the size of apples. By using color image segmentation and OpenCV library, apples of smaller dimensions are recognized. To count the apples inside the box, the Hough transform for circles was tested. Summing recognized apples (outputs) provided by artificial neural networks proved to be a better choice for counting the apples inside the box. The tested algorithms can be used to control the quality of fruit inside boxes in industrial plants.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:612252
Study programmeTitle: Graduate University Study Programme in Computer Engineering
Study programme type: university
Study level: graduate
Academic / professional title: magistar/magistra inženjer/inženjerka računarstva (magistar/magistra inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2023-10-13 09:48:26