Student information

MSc thesis topic: Identifying potato disease in UAV imagery with deep learning

Potato is an attractive crop due to its high yield potential and nutritional benefits. However, it is estimated that approximately 32 percent of potatoes are lost because of diseases. UAV remote-sensing technology, as a non-destructive method, is promising for identifying diseased plants in the field due to its capacity to monitor responses to the biophysical and functional properties of plants. Much research has been done with UAV-based imagery to identify diseased crops with machine learning; however, the identification accuracy heavily relies on manual or well-defined features. In contrast, deep learning could extract the features or attributes from raw data without any human input.

How will a deep learning model be developed to achieve high accuracy in disease identification? This project aims to deploy or design a neural network to identify the diseased plants in the filed using UAV-based imagery. Various of data source, such as RGB imagery, hyperspectral imagery, and LiDAR data, will be available to perform potato disease identification.

Objectives and Research questions

  • Conduct a literature review on disease detection using deep learning methods.
  • Develop or implement a neural network for identifying diseased plants in the field.
  • Evaluate the results and perform a comprehensive analysis of the outcomes.

Requirements

  • Option to conduct fieldwork: needs discussion with supervisors.
  • Experience and/or interest in processing hyperspectral imagery.
  • Good knowledge in scripting, preferably using Python.

Literature and information

  • Kouadio, L., El Jarroudi, M., Belabess, Z., Laasli, S.E., Roni, M.Z.K., Amine, I.D.I., Mokhtari, N., Mokrini, F., Junk, J. and Lahlali, R., 2023. A Review on UAV-Based Applications for Plant Disease Detection and Monitoring. Remote Sensing, 15(17), p.4273.
  • Wieme, J., Leroux, S., Cool, S.R., Van Beek, J., Pieters, J.G. and Maes, W.H., 2024. Ultra-high-resolution UAV-imaging and supervised deep learning for accurate detection of Alternaria solani in potato fields. Frontiers in Plant Science, 15, p.1206998.
  • Gibson-Poole, S., Humphris, S., Toth, I. and Hamilton, A., 2017. Identification of the onset of disease within a potato crop using a UAV equipped with un-modified and modified commercial off-the-shelf digital cameras. Advances in animal biosciences, 8(2), pp.812-816.
  • Polder, G., Blok, P.M., De Villiers, H.A., Van der Wolf, J.M. and Kamp, J., 2019. Potato virus Y detection in seed potatoes using deep learning on hyperspectral images. Frontiers in plant science, 10, p.434052.
  • Franceschini, M.H.D., Bartholomeus, H., Van Apeldoorn, D.F., Suomalainen, J. and Kooistra, L., 2019. Feasibility of unmanned aerial vehicle optical imagery for early detection and severity assessment of late blight in potato. Remote Sensing, 11(3), p.224.

Theme(s): Sensing & measuring, Modelling & visualisation