Industrial process tomography

There is growing interest in using real-time tomography to monitor complex industrial processes. It is often important to know what is happening within a pipe, a vat, or a reaction vessel, and tomography provides a convenient and non-invasive way of doing this. We are particularly interested in novel techniques, such as tomography based on the electrical or magnetic properties of an object. Please contact Manuchehr Soleimani for potential collaboration in these imaging techniques.

Electrical and magnetic tomography

For industrial tomography, the emphasis is usually on high speed data acquisition, rather than resolution. This lends itself to electrical tomography techniques, which although lacking the resolution of X-ray and MRI tomography, are nearly instantaneous. Examples of electrical tomographic imaging methods are:

Electrical Impedance Tomography (EIT), in which measurements of resistance between different combinations of electrodes, are used to determine the internal resistivity of an object.

Electrical Capacitance Tomography (ECT), in which multiple measurements of capacitance between different combinations of metal plates placed around the object, are used to determine the internal dielectric permittivity of an object.

Magnetic Permeability Tomography (MPT), in which multiple measurements of coupling between different combinations of magnets and magnetometers are used to determine the internal magnetic properties of an object.

Magnetic Induction Tomography (MIT), in which multiple measurements of coupling between magnetic excitation and sensing coils are used to determine the internal resistivity of an object.

Object (left) imaged (right) with electrical capacitance tomography.

Ultrasound Tomography

In Ultrasound Tomography (UST), pulses of very high frequency sound (typically in the order of megahertz) are fired into the object to be observed. Ultrasound detectors are used to measure the time-of-flight to different locations on the objects surface. This time-of-flight depends on the physical properties of the intervening material, and so contains information about the object's interior. Multiple measurements can be used to reconstruct the object's internal composition. We are developing a UST device to measure the internal composition of a fluid filled vessel. It can measure velocity throughout the fluid, and detect the presence of solid objects, or gas bubbles.

Tomography Hardware

We are currently constructing a range of imaging devices, many of which are based on these principles. See the Engineering tomography lab web page for more information.

Prototype magnetic induction tomography device.

Tomography software

We have developed the tomography software for ECT, MIT, EIT, Ultrasound tomography and cone-beam CT all in 2D and 3D.

As an example electrical tomography is a difficult mathematical problem, as the inverse problem is nonlinear. Therefore, much of the research in this field is still concerned with developing usable algorithms. Noise is a major problem, as it can be interpreted as the presence of physical features which aren't actually there. In the example below, EIT measurements of test objects (shown left) were simulated, and noise was added to the data. Using conventional algorithms (shown centre), this noise gives the strong impression of other features. With a newer algorithm, these artefacts are greatly reduced.

Comparison of algorithms for electrical impedance tomography.