DEPARTMENT OF ACTUARIAL MATHEMATICS & STATISTICS (AM&S), SCHOOL OF
MATHEMATICAL & COMPUTER SCIENCES (MACS), HERIOT-WATT UNIVERSITY
(with Tullis Russell Papermakers, Markinch, Glenrothes, and
Biomathematics & Statistics Scotland, Edinburgh)
EPSRC CASE STUDENTSHIP (PhD - 3.5 years funding)
Statistical modelling and analysis of spatial processes in
anti-counterfeiting systems
Applications are invited from UK/EU candidates who have, or expect to
have, a good Honours/masters degree in mathematics or statistics to
undertake the above 3.5-year research project commencing 1 October 2005,
under the supervision of Professor Gavin Gibson (AM&S) and Professor
Chris Glasbey (Biomathematics and Statistics Scotland). Project
funding covers all tuition fees, and includes annual maintenance
payments to the student in the form of an EPSRC stipend payable to UK
candidates (£12000 in 2005/6 and rising in later years) plus a
contribution of £2000 per annum from the industrial partners, Tullis
Russell Papermakers. Generous funds for equipment and travel are also
available.
The project will investigate problems in statistical image analysis
arising in anti-counterfeiting systems that use random distributions of
fibres in paper to 'fingerprint' cheques, bonds and other high-value
items. It will involve extensive interaction with Tullis-Russell
Papermakers, who are world leaders in the development of
anti-counterfeiting systems, and will supply data and extensive
knowledge of paper-making technology to the project. A wide range of
mathematical modelling/statistical problems will be considered. These
include: the statistical modelling of fibre deposition in the paper
manufacturing process; parsimonious feature extraction from digitised
images; modelling degradation of items through folding and other
processes; the design of authentication algorithms. The project offers
the opportunity to work with techniques at the forefront of statistical
image analysis including: point process models, spatio-temporal
modelling methods, Bayesian inference and Markov chain Monte Carlo;
wavelets and related transforms. It will also provide the student with
valuable industrial experience and the chance to make an important
contribution to the advancement of a rapidly developing technology of
worldwide importance.
The student will be based in the Department of Actuarial Mathematics and
Statistics (rated 5 in RAE2001)on Heriot-Watt's Riccarton campus. They
will spend extended periods working on-site at Tullis-Russell's paper
mill in Glenrothes and working with Professor Glasbey at BioSS's
headquarters on Edinburgh's King's Buildings campus.
For further information and an application form please contact Gavin
Gibson, Actuarial Mathematics and Statistics, Heriot-Watt University,
Edinburgh EH10 4BY. Tel. +44 131 451 3205, e-mail: [log in to unmask]
The studentship will remain open until it is filled.
Heriot-Watt is an equal opportunity employer.
|