Use Of Image Reconstruction Algorithms Research Articles

Capacitive Multielectrode Direct-Imaging Sensor for the Visualization of Two-Phase Flows

In this paper, a novel capacitive array–sensor to visualize two-phase flow by measuring liquid holdup in pipe cross-section is introduced. The measuring principle is based on the difference between electrical permittivity of the phases. The sensor consists of two copper rings being an emitter and one receiver ring. The receiver ring is divided into segments, which are individually interrogated by the measuring electronics in a fast manner. In this way, flow images are directly generated from acquired signals of electrodes in a way that it visually represents the investigated flow, avoiding the use of image reconstruction algorithms as usual in tomographic techniques. The sensor is fabricated in a flexible printed-circuit board, which is flush-mounted in the inner pipe wall. A measuring electronics is responsible to detect the capacitance changes in the electrodes. The introduced sensor along with the associated electronics has been tested in static and dynamic two-phase flow, both horizontally and vertically. Direct images were generated in these different situations, showing satisfactory results when compared with a reference wire-mesh sensor.

Simulative Study of Two-Phase Homogenous and Isotropic Media Imaging using Magnetic Induction Tomography

Magnetic induction tomography is a new non-invasive technology, based on eddy current discovery of electromagnetic induction by Michael Faraday. Through this technique, the passive electrical properties distribution of an object can be obtained by the use of image reconstruction algorithm implemented in this system. There are many types of image reconstruction that have been developed for this modality, however in this paper only two algorithms discussed, Linear Back Projection and Eminent Pixel Reconstruction. Linear Back Projection algorithm is the most basic type of image reconstruction. It is the simplest and fast algorithm out of all types of algorithms, whereas Eminent Pixel Reconstruction algorithm is an improved algorithm which provided better images and has been implemented in other modalities such as optical tomography. This paper has implemented Eminent Pixel Reconstruction in magnetic induction tomography applications and the performance is compared to Linear Back Projection based on the simulation of the fourteen types of simulated phantoms of homogenous and isotropic conductivity property. It was found that Eminent Pixel Reconstruction has produced better images relative to Linear Back Projection, however the images are still poor when the objects are located near to the excitation coil or sensor and it is worse when the distance between objects are near to each other.

Use of image reconstruction algorithms based on the integral radon transform in small-angle X-ray computer tomography

Use of image reconstruction algorithms based on the integral radon transform in small-angle X-ray computer tomography