The current research students (MSc by Research, MPhil/DPhil) in the Applied Computing department are working in a large variety of different subjects.
If you are interested in more information about studying on a research degree at Buckingham, please contact the Applied Computing Department (by email: email@example.com or phone: +44 (0)1280 828 322) or visit our Postgraduate Study page.
X-ray biomedical imaging processing for automatic spine curvature estimation from X-ray images of a mouse model.
Decision support systems for medical diagnostic decision making: Towards confidence-centric classification based on Gaussian models and Bayesian principle
Complex event processing for decision support systems.
This research investigates the velocity aspects of big data, designs and evaluates more effective algorithms in coping with dynamically changing data sets.
Automatic processing and analysis of high throughput skin images.
This project investigates effective image analysis and classification techniques for automatic identification of signs of miscarriages in early pregnancies using ultrasound images.
Develop simple cryptographic tools that enable dynamic implementation of existing ciphers. Previously was: Research is in the field of cryptography with an interest in developing dynamically changing cipher components e.g. S-boxes.
Enhanced crowd control based on smartphones.
Biometrics systems includes several processes, each of which affects the final outcome of the system which is to accurately identify or authenticate a person. Typical processes of a biometrics system are: quality assessment; pre-processing; feature extraction and classification. Depending on the type of biometric feature used and the operational context of the system, one could adaptively select the most appropriate techniques for each of the above processes to optimise the final outcome of the system. This research will investigate the use of software agents and game theory to develop a framework for a context-aware adaptive biometrics system.
This research explores the essential natures of unstructured data, investigates effective non-relational techniques for modelling and querying unstructured data in databases, and attempts to apply suitable techniques for modelling unstructured data in an example medical school student record database and observe their performances. The aim is to understand how effective existing techniques are in terms of expressing and handling unstructured data in a real application scenario.
This research explores important issues relating to data encryption in real-life databases. The research starts with a general exploration of database security, and then further investigates existing technologies, solutions and systems that can ensure protection of sensitive data through encryption on the one hand and the deterioration of query performance and inconvenience of use on the other together with the trade-off issues between the two aspects. This study in particular scrutinises the CryptDB technology, its strengths and limitations in comparing to other existing solutions, and evaluate its effectiveness when deployed in real-life application systems using the medical school database as a test bed for the evaluation.
This project is to combine security data devised for Adhoc network route discovery protocol with actual data communication security. The work is based on the Prime-IP model, to investigate this can be enhanced further to achieve this and to include location as well.
This project investigates and develops effective and novel computer based solutions in automatically classifying ovarian tumours based on domain specific and image specific features extracted from ultrasound images of ovarian masses.
Application of standard numerical techniques to solve non-linear PDEs such as those that govern curvature of surfaces and higher dimensional Kähler manifolds, or govern to certain image inpainting task.
Comparative study of the effect of various Dimension Reduction techniques for Big Data Analysis.
Automatic processing and analysis of microscopic brain images.
The geometry and topology of quasi-Einstein metrics.
Syeda Fariha Hasnain
Using fast media access strategy (maybe involving game theory) for network access in Cognitive Radio Networks.
Modern smartphones and tablet devices uses ‘pattern unlock’ techniques as an alternative to passwords and PINs to gain access to device functionality. In pattern unlock, users perform a gesture swipe connecting set of dots in a display on the device’s touch screen. The authentication is based on the order in which the dots are connected, which is predefined by each user. This research will evaluate the security of gesture-based pattern unlock mechanism. This will then be followed by the incorporation of dynamic features of touch gestures such as gesture speed, finger pressure, finger movement-time, and gesture accuracy to enhance the security of gesture-based pattern unlock techniques used in smartphones. In addition to these dynamic features, the project will also consider the use of data from smartphone’s accelerometer and/or the gyroscope to enhance the authentication.
Automatic analysis of cellular signalling parameters based on fluorescent video imaging.
This project investigates and develops effective and novel techniques to improve security and privacy in cloud computing.
The aim of the research is to investigate and propose novel mathematical solutions to improve biological networks modelling and pathway analysis, which can help biologists to gain better understanding and interpretation of extremely complex biological data.
Surface registration and Topological Data Analysis for Big Data.
Enhancement of IoT based WSN in terms of authentication and transfer quality.
Enhancement of smartphone based cloud computing services based using off-loading and virtualisation.
Enhancement of contactless cards authentication based on WSN-IoT localisation indoors.
Bank Transactions authentication and security.