Ciência sem Fronteiras (Science Without Borders)
DCU PhD Project Template:
**Please use one form per project**
Please complete & submit to graduate.research@dcu.ie by Friday 13th July
PI name & contact details:
Dr Cathal Gurrin
School:
Computing
Has project been agreed with head (or
nominee) of proposed registration school?
Yes
Research Centre / group affiliation:
CLARITY: Centre for Sensor Web Technologies
Research group / centre website:
http://www.clarity-centre.org
PI website / link to CV:
http://www.computing.dcu.ie/~cgurrin/
Brief summary of PI research / research group / centre activity (2 or 3 lines max):
CLARITY is an SFI-funded CSET which has been in operation for 4 years, since July 2008. It is a
partnership between UCD, DCU and the Tyndall National Institute in Cork and addresses the
emerging world of what is called the sensor web. CLARITY develops its own sensor technologies as
well as using off-the-shelf sensors for applications in areas such as sports, health and wellness,
environment, energy monitoring, ambient assisted living and the online world of social media and
search.
Title & brief description of PhD project (suitable for publication on web):
Using a Geographical Ontology for image classification and object recognition
The image classification and object recognition methods are currently attending an advanced
level of accuracy for domain specific tasks. However, having a reliable and trustable accuracy in
generic tasks is still a research challenge. Relying only the visual information to have a perfect
accuracy seems to be an impossible task due to the lack of other information helping in
identifying objects and scenes. Using an external source of information, can lead to a solution
for the limited effectivness of the "classic" visual classification and recognition methods.
The external source that will be built and explored in this PhD study is a geographical based
Ontology. This Ontology will provide information about the objects and the scenes related to a
specific location (GPS, Google map location, Network location... ).
For example, through this ontology we can answer question like: "What are the objects and the
scenes that are the most likely to find in pictures taken in this place". Using this kind of
information can help the classification method in excluding some object and scenes, or in
making higher the probability of identifying some object or scene.
The study should cover the following points:
1- The construction of such Ontology (Manual, Automatic, semi-automatic)
2- How to explore the Ontology and perform inference on it.
3- Extending the classical sate-of-the-art classification and recognition methods, in order to use
external source of information. Probably, propose a standardization of "How a classification
methods can use an external source of information"
4- Experimentation on standard images collections (Pascal VOC, TRECVid...), the experiments
should focus on the benefits of using the Ontology and how the Ontology leads to better
classification and recognition methods.
Unique selling points of PhD project in DCU:
DCU projects should offer something that’s not available in Brazil – specific equipment, multi-disciplinarity,
aspects of structured programme, links with industry, placements, links with other research groups etc.
CLARITY is a truly inter-disciplinary research centre and within DCU, CLARITY research is carried out in the
schools of Computing, Electronic Engineering, Mechanical Engineering, Chemical Sciences, Health
and Human Performance, Nursing and Human Science, and the Institute of Ethics but all in the one
location on campus. As well as carrying out basic and fundamental research, CLARITY has funded
projects with 57 different industry partners, from large multi-national companies in Ireland like Intel
and Alcatel-Lucent, to international companies like Disney Research, Under Armour and Adidas, to
Irish SMEs like Netwatch, Episensor and Beta Electronics. In a recent independent review of CLARITY
conducted by Science Foundation Ireland, CLARITY was deemed to be "operating as a world class
research centre".
Name & contact details for project queries, if different from PI named above:
Please indicate the graduates of which disciplines that should apply:
Computer Science, Computer Engineering
Ciência sem Fronteiras / Science Without Borders Priority Area:
Please indicate the specific programme priority area under which the proposed PhD project fits- choose only
one (tick box):
Engineering and other technological areas
Pure and Natural Sciences (e.g. mathematics, physics, chemistry)
Health and Biomedical Sciences
Information and Communication Technologies (ICTs)
X
Aerospace
Pharmaceuticals
Oil, Gas and Coal
Renewable Energy
Minerals
Biotechnology
Nanotechnology and New Materials
Technology of prevention and remediation of natural disasters
Biodiversity and Bioprospection
Marine Sciences
Creative Industry
New technologies in constructive engineering
Please complete