The University of East Anglia’s School of Computing Sciences is a world-leading centre for research in the broad field of computer vision, which has many applications in the Agri-food sector, from field and fisheries to fork.
UEA and the Earlham Institute have a long history of successful collaborations in the Agri-food sector, including with BASF, Buhler Sortex, Cefas, Gardline, John Innes Centre, Rothamsted Research, and Syngenta.
The supervisors listed below will be pleased to hear from applicants interested in computer vision and related subjects, within the wider field of Agri-food robotics.
Beatriz is a Data Scientist with experience of developing algorithms and their application to varied environments . She is interested in classification, clustering and rule induction algorithms with particular application to text mining, and complex mixed data including images, text and structured data. She has also worked on uncertainty by looking at how to best deal with missing data in clustering and classification analysis. Application areas have included health (collaborating with members of Norwich Medical School and the Norfolk and Norwich University Hospital), Emergency Preparedness and response (collaboration with Public Health England and King's College London), financial (AVIVA) and others.
She is working currently on an agri-food project developing counting algorithms for complex in-field wheat images in collaboration with the Earlham Institute. She would like to hear from students with an interest in developing algorithms for complex data with particular application to the agri-food sector.
Graham conducts world-leading research in the broad field of “computing how we see”, with strong interests in computer vision and colour imaging. He has a passion for developing algorithms that deploy in the real world, and has a long and successful track record of working with industry and commercialising his research.
With respect to Agri-food Robotics, Graham has had a long collaboration in the sector with Buhler Sortex. His work on shadow removal – e.g. to better support in the field autonomous navigation – deploys in robotic systems. He has a current collaboration with Gardline on understanding the content of underwater images captured by remotely operated vehicles.
Graham would be interested in hearing from PhD candidates with an interest in helping robots see more and see more clearly. Budding entrepreneurs are also encouraged to speak to Graham.
Dr Gong has research interests in the convergence of colour vision, machine (deep) learning and 3-D machine vision. Recently, he has been focused on vision-based methods for understanding real-world illumination and 3-D geometry. In the domain of Agri-Food Robotics, Dr Gong would like to hear from students with an interest in real-time fruit colour processing/recognition or plant visual recognition/segmentation.
Further information about Dr Gong’s interests can be found on his homepage.
I hold a first degree in electronics and a PhD in the application of statistical estimation theory to the passive synthetic aperture problem. My early career was concerned with problems in signal processing, acoustics and vibration. When I joined UEA I was able to apply these interests to image processing and have been working in the field of Computer Vision and Artificial Intelligence since then. In conjunction with Andrew Bangham I have developed a series of novel image analysis algorithms called sieves. I am currently looking at applying these to a number of problems, including automatic lip reading.
I have acted as consultant for a number of international companies and as Executive and Non-Executive Director for several spin-outs and start-ups.
Michal’s interests are in colour vision science, physics based vision, and machine learning. His research has found applications in industries including fisheries and food manufacturing, as well as in nuclear power and the environmental sciences, through PhD research funded by the Innovate UK and NeXUSS Centre for Doctoral Training. His current industrial collaborators include the Centre for Environment, Fisheries and Aquaculture Science (Cefas), Marine Scotland and British Antarctic Survey.
Michal would be pleased to hear from students with research interests similar to his own.
Sarah leads the Digital Humans Group at UEA and her previous work has concerned the analysis and synthesis of faces and bodies during speech. She has worked on computer vision problems such as face and body tracking in video, and machine learning projects such as computer lip-reading, automatic redubbing of video and speech-driven facial animation.
Sarah would be interested in working with students on projects relating to time-series regression or classification, and more generally, machine learning and computer vision in the field of Agri-Food Robotics.
Dr Wang received his BEng (1982) and MEng (1985) degrees from NEU (NorthEastern University, China) in Automatic Control Engineering, and PhD degree in Advanced Computing in 1996 from the University of Manchester Institute of Science and Technology(UMIST), UK.
He and his PhD students conduct research in the areas of data mining/knowledge discovery, ensemble approach and artificial intelligence. He welcomes and encourages any potential applicants to contact him to discuss possible research topics in the context of Agri-tech Robotics.
Tahmina holds a first degree and a PhD in Electrical and Electronic Engineering with research expertise in Digital image and signal processing, Deep learning, Edge computing and Risk prediction modelling.
With respect to Agri-food Robotics, Tahmina’s focus is in developing autonomous agri-robot or drone based vision systems for detection, segmentation (crops versus weeds), classification and tracking of objects such as fruits, plants, food, livestock, people, etc. Vision-based tasks for crop monitoring may include to classify/identify when individual plants are ready for harvest or to detect the onset of diseases.
Tahmina would be pleased to hear from students with research interests similar to the topics mentioned above or along the lines of reinforcement learning based on-device control system development for precision-robotics.