Primerjava algoritmov za registracijo 3D oblakov točk

diplomsko delo

Aljaž Mohorič (Author), Bogdan Lipuš (Mentor)

Abstract

Diplomsko delo je namenjeno primerjavi različic iterativnega algoritma najbližje točke in raziskovanju algoritma in njegovih različic. V nadaljevanju bomo predstavili začetke algoritma, njegovo delovanje, pomembne pojme ter delovanje in teoretično ozadje različic algoritma. Prikazali bomo tudi implementacijo štirih različic algoritma, ki so: iterativni algoritem najbližje točke, iterativni algoritem najbližje točke z uporabo normal, nelinearen iterativni algoritem najbližje točke in posplošen iterativni algoritem najbližje točke. Vse različice bomo testirali z različnimi parametri nad različnimi 3D oblaki točk. Rezultati kažejo, da sta algoritem najbližje točke in algoritem najbližje točke z uporabo normal najboljša za splošno uporabo.

Keywords

diplomska dela;algoritem;točke;registracija oblakov točk;3D oblak točk;diplomske naloge;

Data

Language:	Slovenian
Year of publishing:	2022
Typology:	2.11 - Undergraduate Thesis
Organization:	UM FERI - Faculty of Electrical Engineering and Computer Science
Publisher:	[A. Mohorič]
UDC:	004.92.021(043.2)
COBISS:	99843843
Views:	303
Downloads:	77
Average score:	0 (0 votes)
Metadata:

Other data

Secondary language:	English
Secondary title:	Comparison of algorithms for 3D point cloud registration
Secondary abstract:	In our thesis we study and compare algorithms for aligning 3D point clouds called iterative closest point algorithms and different variants of it. First we explore the beginnings of the algorithm, the theory behind algorithm, important concepts and diffrent variants of the algorithm. We will get to know four variants of algorithm, which are: iterative closest point algorithm, iterative closest point algorithm with normals, nonlinear iterative closest point algorithm and generalised iterative closest point algorithm. All of them will be tested with diffrent parameters and diffrent 3D point clouds. The results show, that iterative closest point algorithm and iterative closest point algorithm with normals are the best in most cases.
Secondary keywords:	thesis;algorithm;points;registration of point clouds;3D point cloud;
Type (COBISS):	Bachelor thesis/paper
Thesis comment:	Univ. v Mariboru, Fak. za elektrotehniko, računalništvo in informatiko, Računalništvo in informacijske tehnologije
Pages:	1 spletni vir (1 datoteka PDF (X, 56 f.))
ID:	14209640