Izdvajanje elementa slike

undergraduate thesis

• Title (english): Extracting elements of image
• Author: Robert Čupković
• Mentor: Dragan Jevtić (mentor)
• Committee member: Dragan Jevtić (predsjednik povjerenstva)
• Committee member: Aleksandar Antonić (član povjerenstva)
• Committee member: Darko Štriga (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2017-07-10, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Electrical Engineering

Abstract

Porastom važnosti slike u životu pojedinca, raste važnost računalnog vida, a time i izdvajanja elementa slike. U tome nam je od velike pomoći biblioteka otvorenog programskog koda OpenCV koja sadržava funkcije i algoritme računalnog vida. Kao moguća rješenja za izdvajanjem elementa slike pokazali su se algoritam GrabCut, kombinacija CannyEdge detekcije i algoritma izdvajanja kontura te uklanjanje pozadine. GrabCut se može implementirati na dva načina te je prilagođen pravokutnim objektima.... More CannyEdge detekcija i izdvajanje kontura izdvajaju sve objekte koje pronađu na slici. Uklanjanje pozadine se koristi na slikama gdje je lako uočiti razliku između prednjeg plana i stražnjeg plana. Kombinacijom SURF-a, detektora i deskriptora svojstva objekta, te GrabCut-a dobiva se vrlo efikasno rješenje uz minimalnu interakciju s korisnikom. Izdvajanje objekta kojeg preklapaju drugi objekti predstavlja područje na kojem je moguć napredak u posljednjem rješenju. Less

Abstract (english)

As the importance of pictures increases in one's life, the importance of computer vision increases as well, and along with that the importance of extracting elements of image. OpenCV, an open-source library containing computer vision algorithms and functions, offers big help in that regard. Possible solutions for extracting elements of image include the GrabCut algorithm, a combination of the CannyEdge algorithm and the contour extraction algorithm and background removal. GrabCut can be... More implemented in two ways and is tailor-made for rectangular objects. The CannyEdge detection and contour extraction solution extracts all objects found on the image. Background removal is used in images where the difference between foreground and background can easily be spotted. A combination of SURF, a feature detector and descriptor, and GrabCut gives us a very efficient solution for extracting elements of an image with minimal user-application interaction. Extracting an element overlaid by other objects is a possible area of improvement in the last solution. Less

• Keywords: Računalni vid; OpenCV; GrabCut algoritam; CannyEdge algoritam; kontura; uklanjanje pozadine; SURF homografija
• Keywords (english): Computer vision; OpenCV; GrabCut algorithm; CannyEdge algorithm; contour; background removal; SURF homography
• Language: croatian
• URN:NBN: urn:nbn:hr:168:454992
• Study programme: Title: Computing; 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 resource: Text
• File origin: Born digital
• Access conditions: Closed access
• Created on: 2019-03-13 19:49:02