Professor Ivan Tyukin (Mathematics)
Professor Kevin Tansey (Geography)
Ref: LIH533 MPhil
The successful candidate will work on a research and development project with Bluesky International Ltd and produce a a thesis for the degree of MPhil.
The characterisation of tree structural features is important for ecological, climate change with respect to changes in above-ground biomass and environmental studies. The University of Leicester is interested in using high point density LiDAR data sets to understand how we can extract meaningful information about the properties of trees using deep learning methodologies within an AI framework.
Bluesky International Ltd, based in Leicestershire, is a leading supplier of aerial survey, GIS and location-based data and services and has offices in the UK, US, Ireland and India. The company is a key partner of the Manufacturing, Engineering, Technology and Earth Observation Research Centre (METEOR) of the University, based at Space Park Leicester. Amongst other services, the company delivers nationwide individual tree mapping to key clients, e.g. Ordnance Survey. In view of the “Flying Cities” campaign Bluesky has recently installed on its airplanes one of the most advanced LiDAR sensor suites on the market, capable of acquiring up to 100 points per square meter.
Bluesky has identified a revenue opportunity in the provision of enriched maps of individual trees including tree structural measurements and species classification. However, the associated research for the development of automated algorithms for the processing of high point density LIDAR datasets requires expertise in data science that the company currently does not have across its staff.
This R&D project will therefore experiment and develop approaches for the processing of the point cloud towards the extraction of tree structural features, with accuracy, speed and confidence. The immediate outputs of this project would be an annotated software code, ideally open source, for tree localisation and structural feature extraction, leading to a synthetic technical documentation of the approach and wider considerations. Of further interest is the detection of objects within the structure of the tree that are temporary features and objects.
The selected candidate will be based at the School of Mathematics, and offered the opportunity to gain industry exposure by executing part of the activities at company premises and at Space Park Leicester.