diplomsko delo
Matic Krepek (Avtor), Simon Klančnik (Mentor), Tadej Peršak (Komentor)

Povzetek

V diplomskem delu je predstavljena uporaba strojnega vida za merjenje razdalje med opazovanim objektom in kamero, s čimer se definira tri prostostne stopnje, ki opišejo položaj predmeta v prostoru oz. njegovo pozicijo. Za popolno definiranje predmeta v prostoru je poleg njegove pozicije potrebna še orientacija, ki je v obravnavanem delu poenostavljena in vnaprej znana. V tem primeru zaznava globine zadostuje za robotsko manipulacijo predmeta. Cilj diplomskega dela je izdelava sistema 3D strojnega vida s kar se da natančno zaznavo razdalje do predmeta. Predstavljena sta dva različna pristopa. Prvi pristop je z uporabo ene kamere in na podlagi eksperimenta ugotoviti funkcijo globine v odvisnosti od površine predmeta na zajeti sliki ter v oddaljenosti središča predmeta od središča zajete slike. Drugi pristop je z dvema kamerama in tako imenovanim stereo vidom, s katerim lahko s pomočjo dveh zajetih slik z dveh različnih vnaprej znanih zornih kotov pridobimo 3D informacije o predmetu. V namen zajema in obdelave slik je bil uporabljen programski paket MATLAB. Uporabljeni sta bili dve spletni kameri Logitech “C920 HD PRO”, ki sta bili vpeti v ohišje, slednje pa je bilo pritrjeno na stojalo. Razdalja med kamerama in objektom se je tekom izvajanja eksperimenta spreminjala. V diplomskem delu so predstavljeni rezultati testiranj obeh razvitih pristopov.

Ključne besede

strojni vid;stereo vid;zaznavanje globine;kamera;MATLAB;

Podatki

Jezik: Slovenski jezik
Leto izida:
Tipologija: 2.11 - Diplomsko delo
Organizacija: UM FS - Fakulteta za strojništvo
Založnik: [M. Krepek]
UDK: 531.717:004.932(043.2)
COBISS: 189945091 Povezava se bo odprla v novem oknu
Št. ogledov: 85
Št. prenosov: 4
Ocena: 0 (0 glasov)
Metapodatki: JSON JSON-RDF JSON-LD TURTLE N-TRIPLES XML RDFA MICRODATA DC-XML DC-RDF RDF

Ostali podatki

Sekundarni jezik: Angleški jezik
Sekundarni naslov: Determining the distance to an object using machine vision
Sekundarni povzetek: The thesis presents the use of machine vision to measure the distance between the observed object and the camera, thereby defining three degrees of freedom that describe the position of the object in space or his position. In order to fully define an object in space, in addition to its position, orientation is also required, which is simplified and known in advance in the work under consideration. In this case, depth perception is sufficient for robotic object manipulation. The goal of the thesis is to create a 3D machine vision system that can accurately detect the distance to an object. Two different approaches are presented. The first approach is to use one camera and based on an experiment to determine the depth function depending on the surface of the object in the captured image and the distance of the center of the object from the center of the captured image. The second approach is with two cameras and so-called stereo vision, with which we can obtain 3D information about the object with the help of two captured images with two different angles known in advance. The MATLAB software package was used for image acquisition and processing. Two Logitech “C920 HD PRO” web cameras were used, which were clamped in the case, and the next one was attached to the stand. The distance between the cameras and the object changed during the experiment. The results of the tests of both developed approaches are presented in the thesis.
Sekundarne ključne besede: machine vision;stereo vision;depth perception;camera;diplomske naloge;
Vrsta dela (COBISS): Diplomsko delo/naloga
Komentar na gradivo: Univ. v Mariboru, Fak. za strojništvo, Proizvodne tehnologije in sistemi
Strani: 1 spletni vir (1 datoteka PDF (IX, 50 f.))
ID: 17936002