Resolution analysis using fully 3D electrical capacitive tomography. (February 2015)

• Record Type: Journal Article
• Authors: Ye, Z.; Wei, H.Y.; Soleimani, M.
• Is Part Of: Measurement. Volume 61(2015:Feb.)
• Page Start: 270
• Page End: 279
• Publication Date: 2015-02
• Subjects: 3D electrical capacitance tomography -- Image resolution analysis -- Volumetric electrical capacitance tomography; Weights and measures -- Periodicals; Measurement -- Periodicals; Measurement; Weights and measures; Periodicals; 530.8
• DOI: 10.1016/j.measurement.2014.10.060
• Languages: English
• ISSNs: 0263-2241

Abstract

Highlights: Fully 3D ECT sensor was developed and experimental results are presented. Limited access 3D ECT tomography is evaluated. Effect of missing data was quantified using three methods of SVD, resolution analysis and image resolution. Quantification of missing sides can become a key design method for 3D ECT. Abstract: Electrical capacitance tomography (ECT) is a non-invasive imaging technique capable of imaging dielectric permittivity contrast of an object. The ECT image reconstruction uses measured capacitance data between pairs of electrode surrounding the object. In some practical ECT applications full access to the surrounding of the object may not be available. In most popular application in ECT for imaging industrial flow through a pipe, there will be missing data from two sides in direction of flow. Another example is a planar array ECT than can have access only to one side of the sample under test. In this paper, a fully 3D array ECT system is developed on six sides of a cubic geometry providing a full access to the imaging region. This fully 3D sensor is used to analyse image quality degradation as a result of missing sides. The sensor development, practical implication, comparison of results with different level of missing data is presented. The resolution analysis is proposed based on theoretical analysis of resolution matrix as well as using singular value decomposition of the Jacobian matrix. The paper presents a number of tools to analyse and quantify the image quality and information that can be achieved in image reconstruction process. The quality of the reconstructed images from experimental data is compared using image resolution parameter for various level of missing data and the results clearly showed how image quality drops with missing information.