About the Project
Compass – the EPSRC Centre for Doctoral Training in Computational Statistics and Data Science – at the University of Bristol in collaboration with FAI Farms are recruiting to a PhD project: A vision-based system for automated poultry welfare assessment through deep learning and Bayesian modelling.
This is an exciting opportunity to join Compass’ 4-year programme with integrated training in the statistical and computational techniques of Data Science. You will be part of a dynamic cohort of PhD researchers hosted in the historic Fry Building, which has recently undergone a £35 million refurbishment as the new home for Bristol’s School of Mathematics.
FAI Farms is a multi-disciplinary team working in partnership with farmers and food companies to provide practical solutions for climate and food security. FAI’s state-of-the-art strategic advice, data insight, and education services, are powered by science, technology and best practice. Our strategic and evidence-based approach is focused on driving meaningful improvements across supply chains, mitigating risks and realising long term business benefits for our partners.
The aim of this PhD project is to create a vision-based system for the automated assessment of chicken welfare for use in poultry farms. The welfare of broiler chickens is a key ethical and economic challenge for the sustainability of chicken meat production. The presentation of natural, positive behaviour is important to ensure a “good life” for livestock species as well as being an expectation for many consumers. At present there are no ways to measure this, with good welfare habitually defined as the absence of negative experience. In addition, automated tracking of individual birds is very challenging due to occlusion and complexity. In this project the student will instead harness and develop novel deep learning approaches that consider individual animals and their behaviours probabilistically within the context of local and general activity within the barn and wider flock. The inferred behaviour rates amongst the flock will then be integrated with on-farm production, health and environmental data through Bayesian time series modelling to identify risk factors for positive welfare, predict farms at risk of poor welfare, and suggest interventions that avoid this scenario.
The PhD will be supervised by statistical data scientist Prof Andrew Dowsey and deep learning experts Dr John Fennell and Dr Laszlo Talas, who together have extensive experience in animal biometrics and agri-tech applications, and animal welfare and veterinary ethics specialist Prof Siobhan Mullan. Industrial co-supervisors at FAI, Annie Rayner and Ralf Onken have long been involved in developing systems for big-data analysis of welfare and monitoring environmental outputs in livestock supply chains. For example, FAI have developed ‘BirdBox’, a laying hen flock management and control system, utilising state-of-the-art sensors controlling light and feed whilst also collecting environmental and production data to provide real-time reports to the farmer.
The studentship would suit a student with a computational or mathematical background, or with a behavioural science or biological background that included a strong computational element. With support from the supervisors the student will be expected to visit partnering farms to collect data and validate the system, but no agricultural expertise or animal handling will be necessary.
Hear from the first two cohorts of COMPASS students at compass.blogs.bristol.ac.uk.
COMPASS and FAI Farms are now recruiting for September 2021 and early application is advised. Interviews with prospective candidates are planned for week commencing 26 April 2021.
To find out how to apply, visit: www.bristol.ac.uk/cdt/compass/apply. Please quote reference number Compass/2021/FAI in the 'Funding' and 'Research Details' sections your application form as well as your Personal Statement. To register your interest or ask any additional questions, please contact the Admissions Team on [Email Address Removed].
Based on your current searches we recommend the following search filters.
Based on your current search criteria we thought you might be interested in these.