Extraction of motion information from electromagnetic induction tomography for flow imaging

Nataša Terzija1, Thanyawee Pengpan2, Anthony J. Peyton1, Wuliang Yin1, Manuchehr Soleimani2 1School of Electrical Engineering, University of Manchester, Manchester M60 1QD, UK,

Natasa.Terzija@manchester.ac.uk

2Electronic and Electrical Engineering, University of Bath, Bath, BA2 7AY UK,

M.Soleimani@bath.ac.uk

ABSTRACT

In this paper, we propose a methodology to extract the motion information from electromagnetic induction tomography (EMT) data. The EMT system used in this study was originally designed for the monitoring application of the molten steel flow through the submerged entry nozzle (SEN) and consists of an array of 8 equally spaced inductive coils arranged around the object, a data acquisition system and a host computer. The present system operates with an excitation frequency 10 kHz and has a capture rate of 10 frames per second.

The movement extraction algorithm has been implemented alongside the image reconstruction software to reconstruction static images and dynamical movement. It is based on calculation of the centre of mass directly from reconstructed images. Results are presented from the implementation of the proposed algorithm on static experimental phantoms, which were moved within the imaging space. The presented results clearly show that algorithm successfully detects the movement of the rigid objects by recovering of the movement of the centre of mass. The other examples shown in this paper actually consider the deformation of the boundaries of the object.

These phantom based experiments shows potential of EMT for visualising flows both in terms of imaging flow and identifying dynamical information.

Keywords electromagnetic induction, tomography, image reconstruction, motion information