Filtering in Image Reconstruction for Electrical Capacitance Tomography

S M Ge1,2 and W Q Yang1

1 Department of Electrical Engineering and Electronics, UMIST, PO Box 88 Manchester M60 1QD, UK, w.yang@umist.ac.uk

2 Academic Visitor from Department of Control Engineering, Harbin Institute of Technology, Harbin 150001, China, shengmin@public.hr.hl.cn

ABSTRACT

The measured data from a tomography system are always contaminated by noise. It deteriorates image quality and may make iterative image reconstruction processes diverge. This paper discusses the use of digital filters in image reconstruction to reduce the noise effect. As a case study, two filters have been implemented for electrical capacitance tomography (ECT), a pre-filter and an in-loop filter. The pre-filter is a 1st-order digital low-pass filter, and the in-loop filter is based on feedback principles. Simulations show that when the filters are used, the images are improved in terms of noise-immunity.

Also it takes less computing time with the filters by using the current image as the initial guess of the

next iteration, providing the possibility of real-time iterative image reconstruction.

Keywords Image reconstruction, Digital filter, Electrical capacitance tomography, Iterative algorithm