Early warning of landslide events using computer vision and geophysical image analysis
Many cuttings and embankments within the UK transport and utilities infrastructure networks were constructed more than 100 years ago and were poorly engineered. Consequently many of these structures are deteriorating and are prone to disruptive failure, particularly during periods of prolonged rainfall, as exemplified by recent extreme weather events. Considerable resources for maintenance and remediation are required, and in some cases earthwork failure can present a risk to life.
A particularly promising earthwork monitoring technology is automated time-lapse electrical resistivity tomography (ALERT), which is a geophysical ground imaging technique. Manual interpretation of ALERT data is prohibitively time consuming, particularly when near-real-time responses to changing ground conditions are required by asset owners to provide early warning of failure events.
We therefore propose to develop automated approaches to analysing these images and data by applying computer vision and machine learning approaches previously developed in medical imaging and/or remote sensing for change detection – thereby providing a foundation on which to integrate the image analyse tools and geophysical instrumentation for industrial slope monitoring.
The project will focus primarily on the development of new computer vision and image analysis tools, but will require the student to work in a multidisciplinary environment with computer scientists, geophysicists, civil and engineers.
The project will build upon recent collaborative research between BGS and the University of Nottingham concerned with automated edge detection within geophysical images, and will utilise geophysical imaging results from previous and ongoing infrastructure monitoring activities. The student will be given training in the processing, modelling and manipulation of geophysical data sets, and will have the opportunity to work at both the British Geological Survey (Nottingham) and the University of Nottingham. The student will also have the opportunity to work with industrial partners (i.e. Network Rail) to ensure that monitoring solutions developed during the project are tailored to the requirements of potential end-users and beneficiaries.
Eligibility: Applicants should hold a minimum of a UK Honours Degree at 2:1 level or equivalent in subjects such as Computer Science, Physics, Engineering, Mathematics or Natural Sciences. For further details contact Dr Jonathan Chambers (email@example.com) or Dr Bai Li (Bai.Li@nottingham.ac.uk)