A selection of recent and ongoing impact generation projects that are underpinned by Computer Science research follow.
Durham Computer Science research on computer vision algorithms enables automated image understanding to provide long-term wide-area surveillance of dynamic scene objects (e.g. people, vehicles) addressing questions such as: “Is there anything there?” (detection); “What is it?” (classification); “Where is it?” (localization); and “What is it’s behaviour?” (tracking). This research, as part the SAPIENT programme, informs scientific work by the governments of UK, USA, Canada, Australia, New Zealand and Netherlands on wide-area, multi-sensor surveillance systems. Our research has contributed to £23.2 million investment in multi-sensor surveillance systems (UK/US government/industry), £11.3 million of additional commercial income to UK companies and supported the creation of around 55 additional science and engineering jobs across six organisations.
As of 2022, the open software architecture developed within the SAPIENT programme is now being used by NATO for the evaluation of counter-drone technologies.
Contact: Toby Breckon
Durham Computer Science research on automatic and algorithmic prohibited item detection, using a range of computer vision techniques on both 2D X-ray and 3D Computed Tomography (CT) imagery, has directly informed UK/US government aviation security policy and provided new enhanced software capabilities for X-ray security scanners across 8 companies who supply the aviation and border security sector. Our work now directly contributes to the security of over 500 million passenger journeys per annum across five continents, with technology from Durham now available at an ever-increasing number of major international airports. The technology has commercial reach to 2-3 billion passenger journeys across 30+ countries globally, and will now help secure all air passengers attending the 2022 FIFA World Cup in Qatar.
Durham's Computer Science research on the use of automated image understanding techniques for future autonomous vehicles (driverless cars) addresses the two key algorithmic tasks within on-vehicle scene understanding: “Where am I?” (known as localization); and “What is around me?” (known as semantic scene understanding). The key challenge is to be able to address these tasks accurately, efficiently (i.e., in real-time relative to the vehicle speed) and robustly under varying environmental (weather) conditions. Our research in this area has directly informed the research and development at two of Europe's leading automotive manufacturers and supported the translation of road vehicle localization technology into rail where it now helps to protect 4.3 billion passenger journeys annually over around 57,000 km of track (Germany/UK).
ExaHyPE is an engine, i.e. a generic collection of state-of-the-art numerical ingredients (high-order time integration, high-order DG representations, block-structured Finite Volume methods, dynamically adaptive Cartesian meshes, task-based load-balancing, and so forth) to solve hyperbolic equation systems given in first-order formulation. The engine allows users to define what (equations) they want to solve, and then it decides how to solve them, where and in which order. It also commits to a particular set of numerical techniques. This degree of freedom on the application domain side in combination with the methodological focus opens the door for various algorithmic optimisations, as all ingredients can be tightly integrated and are aggressively optimised towards each other. Codes using the engine are used to implement solvers simulating various phenomena ranging from gravitational waves to tsunamis and earthquakes.
The code is currently used by multiple companies to assess their software stack and upcoming hardware generations, and it is the backbone of multiple ExCALIBUR research projects in Computer Science.
Contact: Tobias Weinzierl
MammalWeb (www.mammalweb.org) is a citizen-science project established by researchers at Durham University to monitor the UK’s mammals. Camera traps are deployed by members of the public to capture images of wildlife and the resultant images are uploaded to the MammalWeb platform where they are then classified online. MammalWeb’s objectives are to:
The project has collaborated with a range of partners including Durham Wildlife Trust, Great North Museum Hancock, Scottish National Heritage, NatureSpy, British Trust for Ornithology and HMP Deerbolt.
Contact: Steven Bradley
Computer Science has a collaborative project with the NHS University Hospital of North Durham to reduce the ionising radiation exposure of CT scans through utilising deep neural networks. CT scans are expensive in terms of costs and availability, whereas deep generative neural networks are capable of rapidly reconstructing high-quality 3D CT-like images. The success and scalability of generating CT-like images from a small amount of 2D X-ray radiation exposure will have a significant impact on patients (thanks to reduced radiation exposure) and hospitals due to the prohibitively expensive nature of CT scans. Practitioners are obliged by law to consider available alternative techniques which have the same objective but expose patients to less ionising radiation.
Contact: Chris Willcocks
TechUP is a training programme focusing on training individuals from minority groups into tech careers. Working closely with industry, TechUP creates a programme tailored to industry-needs and participant learning experience.
Our most recent programme TechUPWomen took 100 women from the Midlands and North of England, particularly from underrepresented communities: BAME (54%); LGBTQ+ (21%); with disabilities (46%) or dependants (40), with degrees or experience in any subject area, retraining them in technology via a six-month online programme, developed in collaboration with industry, in preparation for roles as software developer, data scientist, agile project manager and business analyst. Our graduates have found new roles or promotions in a wide range of industries: including manufacturing (Jaguar Land Rover, MSP), software (Double Eleven Ltd), education (JISC, Code Nation), service (HR in One) and the public sector (Newcastle City Council, Durham Constabulary). TechUPWomen won the Employment and Skills category in the UK Impact Awards 2020.
Contact Sue Black and Alexandra Cristea
/prod01/prodbucket01/media/durham-university/departments-/computer-science/Image-15---Twizzy-Car-Durham-City.jpg)