prikaz prve stranice dokumenta Segmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata
  Download
PDF 5.13 MB
undergraduate thesis
Segmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata
2017. urn:nbn:hr:200:359977
Ljepić, Srđan
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

Ljepić, S. (2017). Segmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata (Undergraduate 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:200:359977

Ljepić, Srđan. "Segmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata." Undergraduate 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:200:359977

Ljepić, Srđan. "Segmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata." Undergraduate 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:200:359977

Ljepić, S. (2017). 'Segmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata', Undergraduate thesis, Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek, accessed 19 April 2024, https://urn.nsk.hr/urn:nbn:hr:200:359977

Ljepić S. Segmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata [Undergraduate thesis]. Osijek: Josip Juraj Strossmayer University of Osijek, Faculty of Electrical Engineering, Computer Science and Information Technology Osijek; 2017 [cited 2024 April 19] Available at: https://urn.nsk.hr/urn:nbn:hr:200:359977

S. Ljepić, "Segmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata", Undergraduate 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:200:359977

Please login to the repository to save this object to your list.
Metadata
TitleSegmentacija dubinske slike metodom lokalno konveksnih povezanih segmenata
Title (english)Depth image segmentation using Locally Convex Connected Patches method
AuthorSrđan Ljepić
MentorRobert Cupec (mentor)
Committee memberRobert Cupec (predsjednik povjerenstva)
Committee memberZdravko Krpić (član povjerenstva)
Committee memberEmmanuel Karlo Nyarko (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 country2017-09-11, Croatia
Scientific / art field,
discipline and subdiscipline		TECHNICAL SCIENCES
Computing
Artificial Intelligence
Abstract
Opisana je metoda lokalno konveksnih povezanih segmenata, te je ispitana njezina primjena za segmentaciju dubinske slike na objekte. Metoda lokalno konveksnih povezanih segmenata se koristi za segmentaciju slike na osnovu svojstava konveksnosti i konkavnosti objekata. Primjena te metode omogućava dekompoziciju scene na različite objekte što je vrlo važan zadatak robotike. Algoritam se temelji na kreiranju grafa susjedstva supervoksela čiji su čvorovi supervokseli i gdje se promatraju susjedni odnosi između supervoksela koji se mogu promatrati kao konveksne ili konkavne veze. Tada se detektiraju povezani podgrafovi koji predstavljaju segmente povezane konveksnim vezama. Algoritam je izveden koristeći C++ programski jezik u programskom okruženju Microsoft Visual Studio 2013. te PCL biblioteku koja omogućava trodimenzionalnu vizualizaciju slike. Programu su priložene RGB-D slike, a analizom dobivenih rezultata segmentacije moguće je uočiti veliku učinkovitost metode lokalno konveksnih povezanih segmenata u procesu segmentacije slike.
Abstract (english)
A method of locally convex connected patches was described and its application for depth image segmentation is evaluated. The method of locally convex connected patches is used for depth image segmentation based on object convexity or concavity. This method is applicable in robotics and it enables scene decomposition into objects. The considered algorithm creates an adjacency graph of supervoxels, where every patch represents one supervoxel. Edges in the graph are then classified as either convex or concave. Then it detects subgraphs which represent segments. The algorithm is written in C++ programming language and it was implemented in Microsoft Visual Studio 2013. software. PCL library is used for three-dimensional image visualisation. The program was tested using RGB-D images representing the input to the segmentation algorithm. According to the obtained results, the method of locally convex connected patches is shown to be a very effective and reliable
Keywords
Keywords (english)
Languagecroatian
URN:NBNurn:nbn:hr:200:359977
Study programmeTitle: Undergraduate University Study Programme in Computer Engineering
Study programme type: university
Study level: undergraduate
Academic / professional title: sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva (sveučilišni/a prvostupnik/ prvostupnica (baccalaureus/baccalaurea) inženjer/inženjerka računarstva)
Type of resourceText
File originBorn digital
Access conditionsOpen access
Terms of useLicence In copyright icon
Created on2017-11-03 14:43:46