FAST TRACK YOUR CAREER

PHD TITLE: COMPUTER VISION FOR AUTOMATED ASSET DEGRADATION DETECTION

PhD Title:  Video Detection of Asset Degradation

The studentship is with University of Salford and Add Latent Ltd.

Academic Supervisor: Dr Julian Bass

Academic Co-Supervisor: Dr Tarek Gaber

Industrial Supervisor: Hossein Ghavimi

The studentship is fully funded and includes:

  • An MPhil/PhD fee waiver
  • A starting stipend of £18,000 p.a. for four years
  • All bench fees and consumable costs
  • Funds specifically allocated for conference travel

Final date for applications: 14/12/2020

Interviews will be held on: 6/1/2021

The candidate must be in a position to register by 12/2/2021

Description:

Add Latent Ltd. and the University of Salford have been collaborating for over five years. This includes two award winning Knowledge Transfer Partnership (KTP) projects which both received top-scoring certificates of excellence from Innovate UK. Add Energy were also recently awarded the Queens Award for Enterprise in International Trade due to their fast pace of growth.

Computer Science at Salford is ranked in the top 300 departments world-wide (Times Higher Education, World University Ranking, 2021) and Julian Bass was University of Salford, Research Supervisor of the year in 2020 and is Head of the Informatics Research Group.

Add Latent Ltd. have already created an R&D capability using an agile software development process and have an impressive client list of major companies in the industrial marine and energy sector. Their expertise in maintenance optimisation gives them unique access to industrial assets.

This is a great opportunity to join an award-winning research team and to work on a project that will be a world first. The successful candidate will become a thought leader in artificial intelligence and work closely with our clients and company subject matter experts. Previous candidates on the completion of their collaboration projects become key members of the Add Energy R&D team.

This research project aims to:

  • Develop a machine vision platform that uses a five-stage process for 3D model creation, object removal, object detection, semantic correspondence and equipment degradation detection. The platform will be fed with video footage from engineer bodycams recorded over several weeks.
  • Develop and evaluate novel approaches to video detection of equipment degradation which may not be perceptible to the human eye
  • Use novel computer vision techniques to blend video and equipment identification from a library of equipment
  • Implement these novel approaches in software for integration into an existing cloud hosted software infrastructure

These aims will be achieved through three main phases of work. First, establishing a laboratory-based experimental apparatus for testing and evaluating approaches to video detection of asset degradation. Then, you will create novel approaches to video detection of asset degradation. Finally, you will use video processing to recognise asset subsystems and overlay identification information derived from databases.

This project will Implement selected algorithms and approaches to video detection comprising 3D model rendering from video source data, registration of multiple video streams captured on different dates and equipment identification from video.

Candidates:

The successful candidate will work with the Add Latent team to learn about asset integrity, while also developing expertise machine learning and video processing in the university as a member of the Informatics Research Group.

You will hold a Masters degree or minimum of an upper second class honours degree in an area of computer science or software engineering or a closely related discipline, ideally with evidence of previous experience completing an empirical research project. Membership of a relevant professional body would be an advantage, as would experience in video processing, 3D model creation or image processing.

Experience of working with in the commercial digital technology sector would also be an advantage.

As part of your application please provide a CV and covering letter.

Funding Eligibility:

This studentship is only available to students with settled status in the UK, as classified by EPSRC eligibility.  Please visit: https://www.epsrc.ac.uk/skills/students/help/eligibility/

Enquiries: Informal enquiries may be made to Julian Bass by email: j.bass@salford.ac.uk

Curriculum vitae and supporting statement explaining their interest should be sent to m.watts@salford.ac.uk