Computer vision and machine learning for recognition of civil structure components and their damage analysis

   Faculty of Natural Sciences

This project is no longer listed on and may not be available.

Click here to search for PhD studentship opportunities
  Dr B Mandal  Applications accepted all year round  Self-Funded PhD Students Only

About the Project

In this project the PhD research student needs to understand the strengths and weaknesses of the deep convolutional neural network for analysis of civil infrastructures, such as concrete bridges, highways, buildings, tunnels and stadiums, using videos and images. The student research work will be on finding contextual information that can be used for recognition/reidentification of the civil structural components (CSCs), such as columns, beams, slabs, arches, plates, shells, etc. This will be followed by analysis of unhealthy (defects such as spallation, exposed bar, corrosion, crack, etc) areas in the CSCs. The student needs to study, analyse and conduct experiments systematically to model and design new semi-supervised and/or unsupervised deep networks that outperforms the current-state-of-the-art algorithms on standard benchmarking and/or real-world databases. In this project, all databases used will be in accordance to their terms and conditions. Where applicable, appropriate favourable ethics approval will be obtained before any research project studentship offer is made. There is an expectation that the student will also research and apply relevant legal and ethical issues in data collections and analysis in this domain.

These are active and important areas of research with many opportunities for innovation and collaboration. This project will provide an opportunity to pursue world-class research environment, provide experience of design and evaluation processes and an opportunity for substantial contribution to international publication in leading journals/conference/workshop proceedings.

Applications are welcomed from science, technology, engineering or mathematics graduates with (or anticipating) at least a 2.1 honours degree or equivalent. Applicants should have good computing skills and an enthusiasm for designing and testing new algorithms. They should be self-motivated and have the ability to work both independently and as part of a team.

This opportunity is open to UK/EU and overseas students. The collaborative and presentation aspects of the research require good English language and communication skills. Overseas applicants would therefore require an English IELTS (or equivalent) of 6.0 overall with no less than 5.5 in any subtest.


Informal enquiries about the project are very welcome by email to the Project Lead, Dr Bappaditya Mandal ([Email Address Removed]). Full applications should be submitted to:

Please quote FNS 2021-13 on your application.

Keele University values diversity, and is committed to ensuring equality of opportunity. In support of these commitments, Keele University particularly welcomes applications from women and from individuals of black and ethnic minority backgrounds for this post. The School of Computing and Maths and Keele University have Athena Swan Awards and Keele University is a member of the Disability Confident scheme.

More information is available on these web pages:

Funding Notes

Open to fully self-funded full time / part time students only.
Please note that self-funded applicants must provide funding for both tuition fees and living expenses for the 3 year duration of the research. There is a future possibility of competitive scholarship awards for outstanding applicants (1st class honours), however, none are currently available.
For information regarding University tuition fees please see:
Source of funding This opportunity for self-funded applicants only (for example, international students with government or industry sponsorship and UK students with Doctoral Loan funding:
Search Suggestions
Search suggestions

Based on your current searches we recommend the following search filters.

PhD saved successfully
View saved PhDs