Optični tok za identifikacijo nihanj znotraj enega slikovnega elementa

Šonc, Jaša (author), Slavič, Janko (mentor)

V zaključni nalogi je raziskan potencial identifikacije pomikov z amplitudo nižjo od velikosti enega slikovnega elementa. Najprej je gradientna metoda optičnega toka za identifikacijo preizkušena v sintetičnem okolju, nato je izveden realni preizkus s hitro kamero. V programskem okolju Python je razvit programski paket, s katerim so določene optimalne lokacije za identifikacijo, izmerjene lastne frekvence merjenca in njim pripadajoči razmerniki dušenja.

• Language: Slovenian
• Keywords: optični tok, gradientna metoda, hitra kamera, nihanja, Fourierjeva transformacija, lastne frekvence, razmerniki dušenja
• Work type: Final paper
• Typology: 2.11 - Undergraduate Thesis
• Organization: FS - Faculty of Mechanical Engineering
• Place of publishing: Ljubljana
• Publisher: [J. Šonc]
• Year: 2021
• Number of pages: XII, 31, [1] str.
• PID: 20.500.12556/RUL-129174
• UDC: 531.776:534:517.443(043.2)
• COBISS.SI-ID: 81499907
• Publication date in RUL: 28.08.2021

Secondary language

• Language: English
• Title: Identification of subpixel vibrations using optical flow

Abstract

In this final thesis, the potential of identifying displacements with an amplitude smaller than the size of one pixel is investigated. First, the gradient-based optical flow method for displacement identification is tested in a synthetic environment, then a real experiment with a high-speed camera is carried out. Software is developed using Python programming language and is later used for determining the optimal locations for identification and measuring eigenfrequencies and damping ratios of an object.

• Keywords: optical flow, gradient-based method, high-speed camera, vibrations, Fourier transform, eigenfrequencies, damping ratios