A novel method for thinning branching noisy point clouds

Author

Hasirci, Zeynep ; Ozturk, Mehmet

Author_Institution

Dept. of Electr. & Electron. Eng, Karadeniz Tech. Univ., Trabzon, Turkey

fYear

2013

fDate

2-4 July 2013

Firstpage

713

Lastpage

716

Abstract

In this study, we are focused on thinning branching noisy data sets into curves. Robust Line Fitting (RLF) method is proposed for local linear region determination and also overcome problems which occur in high curvature regions. The performance of the RLF method is tested on different noisy branching data sets. These sets are generated artificially in three separation angles (30°, 60°, 90°) and different noise levels (0.1, 0.2, 0.3). To the best of our knowledge, there is no study about non-simple noisy curve reconstruction. Thus, a comparison is made with our previous method which is useful only for simple noisy point clouds. As a result, RLF is an efficient method for not only simple curves but also non-simple curves.

Keywords

curve fitting; signal denoising; signal reconstruction; high curvature region; local linear region determination; noisy branching data set; nonsimple noisy curve reconstruction; robust line fitting method; thinning branching noisy data sets; thinning branching noisy point clouds; Fitting; Image reconstruction; Noise; Noise level; Noise measurement; Polynomials; Robustness; Branching Curve; Curve Reconstruction; Eigenvalue Analysis; Noisy Point Cloud; Robust Line Fitting;

fLanguage

English

Publisher

ieee

Conference_Titel

Telecommunications and Signal Processing (TSP), 2013 36th International Conference on

Conference_Location

Rome

Print_ISBN

978-1-4799-0402-0

Type

conf

DOI

10.1109/TSP.2013.6614030

Filename

6614030