prikaz prve stranice dokumenta Prepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki
  Download
PDF 2.06 MB
master's thesis
Prepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki
2017. urn:nbn:hr:200:958046
Radočaj, Kristijan
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

Cite this document

Radočaj, K. (2017). Prepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki (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:200:958046

Radočaj, Kristijan. "Prepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2017. https://urn.nsk.hr/urn:nbn:hr:200:958046

Radočaj, Kristijan. "Prepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki." Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, 2017. https://urn.nsk.hr/urn:nbn:hr:200:958046

Radočaj, K. (2017). 'Prepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki', Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 07 December 2024, https://urn.nsk.hr/urn:nbn:hr:200:958046

Radočaj K. Prepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki [Master's thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2017 [cited 2024 December 07] Available at: https://urn.nsk.hr/urn:nbn:hr:200:958046

K. Radočaj, "Prepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki", Master's thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, Osijek, 2017. Available at: https://urn.nsk.hr/urn:nbn:hr:200:958046

Please login to the repository to save this object to your list.
Metadata
TitlePrepoznavanje voća na RGB-D slikama na temelju boje i geometrijskih značajki
Title (english)Fruit Recognition in RGB-D Images Based on Color and Geometric Features
AuthorKristijan Radočaj
MentorRobert Cupec (mentor)
MentorIvan Vidović (mentor)
Committee memberRobert Cupec (predsjednik povjerenstva)
Committee memberEmmanuel Karlo Nyarko (član povjerenstva)
Committee memberIvan Vidović (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 country2017-10-02, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Process Computing
Abstract
Razvijen je program za prepoznavanje voća na RGB-D slikama. Cilj programa je testiranje trenutno dostupne tehnologije u primjeni automatizacije branja voća, a rezultati testiranja pokazali su da trenutna tehnologija ima potencijal za korištenje u automatskom branju voća. Ovaj rad se fokusirao na testiranje učinkovitosti dva deskriptora: histograma distribucije normala i SHOT deskriptora. Deskriptori se izračunavaju za svaki segment dobiven segmentacijom RGB-D slike na približno konveksne površine. Klasifikacija segmenata na voće i ostale objekte, koji ne predstavljaju voće, ostvarena je metodom k-najbližih susjeda implementiranom pomoću FLANN biblioteke. Učinkovitosti dviju razvijenih metoda ispitane su na skupu od 2974 RGB-D slika.
Abstract (english)
A program for fruit recognition in RGB-D images is developed. The programs goal was to investigate the practical applicability of the technologies available today for fruit picking automation. Test results have shown that this technology have potential for use in automated fruit picking. The focus of this work is on comparison of two descriptors: normal distribution histogram and SHOT descriptor. The descriptors are calculated for every segment provided by segmentation of RGB-D image to approximately convex surfaces. Classification of the obtained segment on fruits and objects that do not represent fruits was accomplished with the k-nearest neighbors classifier implemented using FLANN library. The efficiency of two the developed methods was tested on a set of 2974 RGB-D images.
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:958046
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 on2017-10-19 13:43:37