Early disease detection using airborne monitoring
Project title:
Artificial intelligence methods for early disease detection using hyperspectral and thermal imagery
Information and objectives:
This project, led by Professor Pablo Zarco-Tejada and Dr Tomas Poblete from the University of Melbourne, focused on developing improved remote sensing methods for the early detection of diseases, including Australia's number one National Priority Plant Pest Xylella fastidiosa (Xf).
Using the university's Airborne Remote Sensing Facility aircraft, thermal and hyperspectral imagery was collected from several thousand hectares of healthy almond, citrus and olive trees in the Mallee region. Artificial intelligence algorithms based on physical models were used to analyse the imagery, avoiding the need for extensive field data collection.
By deriving biophysical and structural parameters from the hyperspectral information, it was possible to develop artificial intelligence (AI) methods coupled with remote sensing technology to derive algorithms for the early detection of plant diseases in Australia, including Xf.
Professor Zarco-Tejada has also been involved in building databases in Spain and Italy, where Xylella has been detected. During 2023, Professor Zarco-Tejada is one of the official partners and Work Package Co-Leader of the European project BeXyl: Beyond Xylella, focused on the development of methods to detect and eradicate the Xf bacteria.
Project funding:
Department of Agriculture, Fisheries and Forestry (formerly Department of Agriculture, Water and Environment)
Project collaborators:
The University of Melbourne, Mallee Regional Innovation Centre, QuantaLab-IAS-CSIC
Status:
Completed (April 2022)
Project lead:
Professor Pablo Zarco-Tejada
Team Leader, HyperSens Laboratory
School of Agriculture, Food and Ecosystem Sciences, Faculty of Science | Department of Infrastructure Engineering, Faculty of Engineering and Information Technology
University of Melbourne
Resources:
- Tackling a global crop pandemic from the air, MRIC newsletter, November 2021
- Divergent abiotic spectral pathways unravel pathogen stress signals across species, Nature Communications, 2021
- Discriminating Xylella fastidiosa from Verticillium dahliae infections in olive trees using thermal- and hyperspectral-based plant traits, ISPRS Journal of Photogrammetry and Remote Sensing, 2021
- Algorithms can help detect Xylella fastidiosa – scientists, Alto Adige, 2021
- Using hyperspectral imagery and a multi-stage machine learning algorithm to distinguish infection symptoms caused by two xylem-limited pathogens presentation slides, 3rd European Conference on Xylella fastidiosa, 2021
- 3rd European Conference on Xylella fastidiosa presentation, YouTube, 2021 (from minute 03:00)
- Projects in the Mallee region on biotic and abiotic stress detection using hyperspectral and thermal imaging, YouTube, June 2020