EPSRC Centre for Doctoral Training in Agri-Food Robotics: AgriFoRwArdS Logo. EPSRC Centre for Doctoral Training in Agri-Food Robotics: AgriFoRwArdS site title logo image.

EPSRC Centre for Doctoral Training in Agri-Food Robotics: AgriFoRwArdS - agri for web

Grey Churchill

  • University of Lincoln in collaboration with Syngenta

Research Interests

Grey’s research interest include, open-source hardware and software.

Publications

Presentations

  • “Automatic Detection of Black Rot in Images of Grapes” (oral) – AgriFoRwArdS CDT Summer School 2021 [June 2021] – Online.
  • “Fish Sorting: A whatsinyour.net solution” (oral) – AgriFoRwArdS CDT Summer School 2022 [July 2022] – Norwich, UK.
  • “Iterative Particle Swarm Optimisation –Hyperparameter Tuning” (poster) – Towards Autonomous Robotic Systems (TAROS) 2023 / AgriFoRwArdS CDT Annual Conference 2023 / Joint Robotics CDT Annual Conference 2023 [September 2023] – Cambridge, UK.
  • “Title Unknown” (poster) – AgriFoRwArdS CDT Summer School: Robotic Phenotyping [July 2024] – Wageningen, The Netherlands.
  • “SLAM (Simultaneous localisation and mapping)” (oral) – AgriFoRwArdS CDT Summer School: Robotic Phenotyping [July 2024] – Wageningen, The Netherlands.

Other Activities

  • Represented the AgriFoRwArdS CDT in the AgriFoRwArdS CDT video.
  • Member of the AgriFoRwArdS CDT Equality, Diversity and Inclusion (EDI) Panel (March 2022 to present).
  • Discussion Panel member at the AgriFoRwArdS CDT Annual Conference 2022 – Discussion topic: AgriFoRwArdS PhD research progress [July 2022].

About me

My long-term goal is to be able to create/modify an open-source system, which is accessible to as many people as possible.

MSc Project

Machine Learning for the Detection of Weeds among Sugar Beets

This project will create a vision system able to detect and localise weeds in images gathered from an RGB camera mounted on phenotyping robots. Previous work has led to a number of systems that can provide bounding boxes of weeds in images, however the accuracy of localisation is a rarely used metric during evaluation. The output of the system proposed by this project in intended for informing the use of herbicides, and as such the localisation accuracy will be key to its success. The data used to train the model(s) will be a combination of labelled images gathered from the University of Lincoln’s Riseholme campus and Campus Klein Altendorf in Bonn, Germany.