Estimation of motion parameters

using 2d lines without correspondences

based on virtual electric potential model

 

B. Chaouki1*, B. Bouda2, Lh. Masmoudi3, D. Aboutajdine2

1 GII Laboratory, National School of Applied Sciences of Agadir, University of Ibn Zohr, BP 1136, Agadir, Morocco

2 GSCM-LRIT Laboratory, Faculty of Sciences, University of Mohamed V, Ibn Battouta Av., BP 1014 RP, Rabat, Morocco

3 LETS Laboratory, Faculty of Sciences, University of Mohamed V, Ibn Battouta Av., BP 1014 RP, Rabat, Morocco

* Corresponding author. E-mail: chaouki@ensa-agadir.ac.ma , chaouki_bk@hotmail.com

Received: 20 February 2007; revised version accepted:13 June 2008

 

Abstract

     Motion estimation is widely used in image processing and computer vision applications. In this paper, we propose new method for estimation of motion parameters using 2D lines without correspondences based on virtual electric potential model, which demonstrate the best results. The basic idea is to model the image as a grid of virtual electric charges of a plane surface in electrostatic balance.

     The algorithm is achieved in two main steps. The first one consists to exploit the corners detected by an improved version of Harris and Stephens detector. Characteristic of gradient vectors at the corners is used in order to draw the strait lines. The second one uses the invariance property of the correlation matrix eigenstructure decomposition. This matrix is formulated from the directing vectors of the straight lines. To determine the correspondence between the lines and to remove the outliers we use an affinity function based on a heuristic criterion.  

     The performances of the method are degraded considerably in presence of noise. For this reason, a pre-processing stage is suitable. It consists to estimate the noise correlation matrix by evaluating iteratively the noise subspace in order to improve the signal noise ratio (SNR). The robustness of the method to the noise and the outliers is remarkable in synthetic or real images. 

 

Keywords: Corner detector; Electric potential; Gradient vector; Line correspondences; Motion estimation; Outliers removal.


© 2015