Projects in the Mallee region on biotic and abiotic stress detection using hyperspectral and thermal imaging

Professor Zarco-Tejada is working on a project funded by the Federal Department of Agriculture, Water and Environment on the Artificial Intelligence methods for early disease detection using hyperspectral and thermal imagery.

This project will focus on Xylella fastidiosa plant pathogen bacterium, which the Department ranks as the number one threat to Australia. Methods need to be developed to detect diseases as fast as possible, and with the capability to cover a large acreage of crop.

The project is part of the preparedness activities to support prevention of such diseases. The focus crops will be almond, olive and table grapes in the Mallee region.

The objective of the project is to develop improved remote sensing methods for the early detection of diseases including the number one national priority plant pest Xylella fastidiosa

Artificial intelligence algorithms based on physical models will be used to analyse hyperspectral and thermal imagery, avoiding the need for extensive field data collections.

Professor Zarco-Tejada is also involved in the build of databases in Spain and Italy where the Xylella has been detected. Data collected is fed into models that can identify hot spots which allow farm managers to take action.

How does an airborne facility work?

Using an airborne facility that can fly over large areas such as thousands of hectares, thermal and hyperspectral imagery can be collected through sensors (hyperspectral and thermal). By using plant and crop traits, the data collected can identify nutrient, water and other stresses.

Practical applications include mapping water stress in precision agriculture. In some cases, stress levels are required to be maintained as they affect the final quality of the crop and yield, helping to produce a more desirable product for market.

The project will use an aeroplane to collect sensory data. Data will also be collected on the ground. Once data is collected it will be analysed and information provided to partners of the project. What sets Professor Zarco-Tejada’s project apart will be the collection of hyperspectral data and the use of a ‘one of a kind’ camera in Australia.


Professor Pablo Zarco-Tejada
Remote Sensing and Precision Agriculture
School of Agriculture and Food, Faculty of Veterinary and Agricultural Sciences and the
Department of Infrastructure Engineering, Faculty of Engineering and Information Technology.
The University of Melbourne