Mjerenje udaljenosti u ravnini ceste kamerom postavljenom na vozilu

undergraduate thesis

• Title (english): Performing measurements in the ground plane by a camera mounted on a vehicle
• Author: Josip Milić
• Mentor: Siniša Šegvić (mentor)
• Committee member: Alan Jović (predsjednik povjerenstva)
• Committee member: Darijan Marčetić (član povjerenstva)
• Committee member: Siniša Šegvić (član povjerenstva)
• Granter: University of Zagreb; Faculty of Electrical Engineering and Computing; Zagreb
• Defense date and country: 2015-07-03, Croatia
• Scientific / art field, discipline and subdiscipline: TECHNICAL SCIENCES; Computing

Abstract

Ovim radom opisani su postupci dobivanja modela stvaranja slike, kalibracije
kamere i preslikavanja točaka slike u ravninu ceste. Tim postupcima ostvarilo se
mjerenje udaljenosti u ravnini ceste uz pomoć slike dobivene kamerom postavljenom
na vozilu. Za mjerenje potrebna je relativno jeftina kamera i tanka podloga s
kalibracijskim uzorkom. Postupak kalibracije kamere implementiran je uz pomoć
C++-a, a ravninsko preslikavanje uz pomoć Pythona. Metode računalnog vida,
potrebne za ovaj rad, implementirane su uz pomoć OpenCV biblioteke. Korišteni
NumPy paket omogućio je stvaranje i korištenje matrica. Eksperimentalni rezultati
pokazali su relativno vrlo precizno mjerenje udaljenosti u ravnini ceste korištenjem
opisanih postupaka. Veće pogreške mjerenja uočile su se kod točaka u ravnini ceste
koje su se nalazile na relativno velikoj udaljenosti od kamere.

Abstract (english)

This work explains obtaining image creation model, camera calibration and
perspective mapping of image points in the ground plane. By that, measurements in the ground plane can be made using images created with camera mounted on a vehicle. Measurements require inexpensive camera and thin base with calibration pattern. Camera calibration procedure was implemented in C++ and planar projection was implemented in Python. Computer vision methods, needed for this work, were implemented using OpenCV library. Imported NumPy package allowed creation and usage of matrices. Experimental results showed measurements in the ground plane with relatively high precision using procedures described in this work. Larger errors were noticed when measurements were performed between points in the ground plane located relatively far from camera.

• Keywords: računalni vid; model stvaranja slike; kamera; kalibracija; uzorak; homografija; ravninsko preslikavanje; mjerenje; udaljenost u cesti; C++; Python; OpenCV; NumPy
• Keywords (english): computer vision; image creation model; camera; calibration; pattern; homography; planar projection; measurements; measurements in the ground plane; C++; Python; OpenCV; NumPy
• Language: croatian
• URN:NBN: urn:nbn:hr:168:205944
• 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: 2018-02-05 17:46:26