The aim of this project is to contribute to the improvement of the ultrasonic tomography techniques for non-destructive inspection and image reconstruction of structures and materials relevant to civil and mechanical engineering. Tomography is the process of imaging a cross-section of some solid. It enables to see through the objects instead of making real cuts, therefore, it has important applications in the medical field and in non-destructive structural health monitoring. The Radon transform and its generalization have played a significant role in the development of various imaging techniques. The ordinary Radon transform has been the foundation of the mathematical model of conventional computed tomography. The inverse transform reconstructs the image of the object onto a plane: to reconstruct a 3D image, several slices are needed.
The main aim of this project is to formulate a novel imaging technique, which employs a 2D Radon transform, that is a projection onto a plane, whose inverse could enable the reproduction of a 3-D image of what is inside the scanned object. The project will cover the theoretical formulation and numerical (finite-elements) simulations to prove the feasibility of the approach proposed. Besides the improvement of the understanding of the inner content of a solid, the novel formulation proposed will require a reduced number of scans needed to perform a 3D image reconstruction in comparison with classical approaches, resulting in shorter acquisition and processing time.
In addition to the improvement of image reconstruction algorithms, one of this project¿s aims is also the development of quantitative tomographic imaging techniques based on coupled physics of waves. In coupled physics imaging, contrast and resolution originating from different physical phenomena are combined. In the project, it is envisaged to employ the thermoacoustic effect in addition to the classical ultrasonic wave response.
Comparing with other work which was done in the past on the same subject, this project exhibit some advantages, which are listed here below:
Novelty: This project proposes a novel imaging technique. It employs a 2D Radon transform, which is a projection onto a plane instead of onto a line. Inverse image reconstruction using projection onto a plane results in the need of a reduced number of projections than when using projections onto a line. Consider the case of the image-reconstruction of a sphere. With the traditional Radon transform, for each slice, two projections as necessary and sufficient. However, the number of slices is theoretically infinite. On the contrary, with the proposed 2D Radon transform, only three projections onto planes will be necessary and sufficient to reconstruct the image of the sphere.
Cost-effective: the novel formulation proposed will require a reduced number of scans to perform a 3D image reconstruction, meaning simpler and faster computer software, with reduction of costs and energy consumption.
Saving of time: the main asset is quick investigation and easy evaluation through out proposed method. The novel technique will require a reduced number of scans, which means less projections and therefore less complex acquisition and processing time. By means of this simplification the whole processing time will be reduced.
Increased resolution: one of the objectives of this project is the improving of the resolution of the image that will be obtained by inverse image reconstruction. By exploiting the thermoacoustic effect and superposing the information deriving from these waves to that obtained from the classical ultrasonic waves, the desired increased resolution will be achieved.
All these aspects are hopefully expected to speed up and contribute to science and further research regarding 3D ultrasonic tomography for non-destructive inspection and image reconstruction. This direction is very important to practitioners who are looking for fast, accurate and high resolution methods in order to get 3D ultrasonic tomography and image reconstruction. The key point of the research proposed is introducing a novel technique that simplifies and fastens the algorithms, but also enables to reach greater resolutions exploiting Multiphysics data.