Enhancing the confidence of segmenetation maps using complimentary data and convolutional segmentation

Accurate maps of the distribution of phases inside multiphase materials is vital for extracting characteristic metrics (such as volume fractions and surface areas), as well as for specifying mesh domains for use in multiphysics simulations. Various high resolution imaging techniques have been developed for this task; however, each has its own strengths and weaknesses. For example, focused ion-beam scanning electing microscopy (FIB-SEM) can achieve very high resolution (c. 5 nm) and phase sensitivity, but struggles to capture representative volumes and is a destructive technique. Alternatively, X-ray computed tomography (XCT) can capture large volumes in-situ, but often performs poorly at differentiating phases, particularly of small features.
By combining image data from multiple techniques (or even just multiple modes or detectors of a single technique) it is possible segment data (i.e. assign each pixel to a phase) with more confidence than any individual method. However, this requires the use of convolutional methods that are able to combine the information from various input channels. In particular, deep convolutional neural networks are very well suited to this kind of task and have seen dramatic advancements in their implementation over the last 5 years due, in part, to their use in self driving cars.
This project would develop machine learning based convolutional segmentation tools for confidently characterising multimodal image data, with potential for impact across all areas of materials science.

The production and processing of materials accounts for 15% of UK GDP and generates exports valued at £50bn annually, with UK materials related industries having a turnover of £197bn/year. It is, therefore, clear that the success of the UK economy is linked to the success of high value materials manufacturing, spanning a broad range of industrial sectors. In order to remain competitive and innovate in these sectors it is necessary to understand fundamental properties and critical processes at a range of length scales and dynamically and link these to the materials' performance. It is in this underpinning space that the CDT-ACM fits.

The impact of the CDT will be wide reaching, encompassing all organisations who research, manufacture or use advanced materials in sectors ranging from energy and transport to healthcare and the environment. Industry will benefit from the supply of highly skilled research scientists and engineers with the training necessary to advance materials development in all of these crucial areas. UK and international research facilities (Diamond, ISIS, ILL etc.) will benefit greatly from the supply of trained researchers who have both in-depth knowledge of advanced characterisation techniques and a broad understanding of materials and their properties. UK academia will benefit from a pipeline of researchers trained in state-of the art techniques in world leading research groups, who will be in prime positions to win prestigious fellowships and lectureships. From a broader perspective, society in general will benefit from the range of planned outreach activities, such as the Mary Rose Trust, the Royal Society Summer Exhibition and visits to schools. These activities will both inform the general public and inspire the next generation of scientists.

The cohort based training offered by the CDT-ACM will provide the next generation of research scientists and engineers who will pioneer new research techniques, design new multi-instrument workflows and advance our knowledge in diverse fields. We will produce 70 highly qualified and skilled researchers who will support the development of new technologies, in for instance the field of electric vehicles, an area of direct relevance to the UK industrial impact strategy.
In summary, the CDT will address a skills gap that has arisen through the rapid development of new characterisation techniques; therefore, it will have a positive impact on industry, research facilities and academia and, consequently, wider society by consolidating and strengthening UK leadership in this field.