A novel method for determination of best ordering direction for noisy point clouds


ÖZTÜRK M., HASIRCI Z.

International Symposium on Innovations in Intelligent Systems and Applications, Trabzon, Turkey, 2 - 04 July 2012, pp.1-5

  • Publication Type: Conference Paper / Full Text
  • City: Trabzon
  • Country: Turkey
  • Page Numbers: pp.1-5
  • Karadeniz Technical University Affiliated: Yes

Abstract

In this paper, we propose a method to determine the 
best regression axis to order the noisy points for curve 
reconstruction. The fundamental problem of the curve 
reconstruction is to order the data in most suitable way. We 
suggested a histogram based feature to be able to determine the 
goodness of the order for a regression line. The method developed 
by using the proposed feature tested on the some synesthetic data 
with different noise levels. This data was selected due to the 
eigenvector approach gave wrong results when obtaining 
regression line. The results showed that the proposed method is 
encouraging.