Analysis and Performance Evaluation of Entropic Thresholding Image Processing Techniques for Electrical Capacitance Tomography Measurement System

Authors

Josiah Nombo, Electronics and Telecommunications Engineering Department, University of Dar es Salaam, TanzaniaFollow
Alfred Mwambela, Electronics and Telecommunications Engineering Department, University of Dar es Salaam, TanzaniaFollow
Michael Kisangiri, Electronics and Telecommunications Engineering Department, Nelson Mandela African Institution of Science and Technology, TanzaniaFollow

Section

Physical Sciences

Abstract

To improve image quality generated from the electrical capacitance tomography measurement system, the use of entropic thresholding techniques is investigated in this article. Based on the analysis of the principle of Electrical Capacitance Tomography (ECT) image reconstruction and entropic thresholding, various algorithms have been proposed for easy extraction of quantitative information from tomograms generated from the ECT system. Experiments indicate that proposed algorithms can provide high-quality images at no or minimum computational cost. It is easier to implement and integrate with classical algorithms such as Linear Back Projection, Singular value decomposition, Tikhonov regularization, and Landweber. Entropic thresholding techniques present a feasible and effective way toward the industrial utilization of ECT measurement systems. Keywords: Electrical Capacitance Tomography; Inverse Problem; Image Reconstruction; Entropic Thresholding;

Recommended Citation

Nombo, Josiah; Mwambela, Alfred; and Kisangiri, Michael (2021) "Analysis and Performance Evaluation of Entropic Thresholding Image Processing Techniques for Electrical Capacitance Tomography Measurement System," Tanzania Journal of Science: Vol. 47: Iss. 3, Article 5.
Available at:https://doi.org/10.4314/tjs.v47i3.5