Early plant disease detection using hyperspectral imaging combined with machine learning and IoT

February 25th, 2019

Leveraging artificial intelligence algorithms to reduce environmental impact and spraying rate of horticultural production.

Plant diseases are responsible for major economic losses in yield and quality affecting agricultural industry worldwide. Disease control strategies are widely focused on spraying pesticides uniformly over cropping areas at different times during the growth cycle. These control strategies, though effective, have adverse economic and ecological effects, introducing new pests and elevating chemical resistance.

Plant disease prediction per pixel.  Left image is the color camera. In the right image, pixels coloured green were classified as control and those coloured red were classified as diseased using a hyperspectral camera. 

In this project*, CSIRO’s team, led by Dr. Peyman Moghadam, is working on developing artificial intelligence algorithms for early plant disease detection using hyperspectral imaging and wireless sensor networks (IoT). Hyperspectral imaging combined with machine learning provides an opportunity to develop fast and non-invasive methods of detecting plant diseases and potentially discriminating between different disease types (e.g.,virus, fungus, bacteria) before the human eye can see them.

According to Dr Moghadam, “the key benefits of the project include early disease identification, reduced use of pesticides, real time precision management such as selective spraying, and real-time continuous decision making to assist growers in better managing inherited risks associated to horticultural production”.

The Robotics and Autonomous System Group is extremely well positioned to deliver this type of project given its experience in the AgTech sector, as well as a highly skilled team which includes world class researchers and engineers.

Having the ability to demonstrate a suite of technologies in the agricultural domain, the Group is open to partnerships and collaborations for research, development, and commercialisation.

For more information, contact Dr Peyman Moghadam.



Moghadam, Peyman; Ward, Daniel; Goan, Ethan; Jayawardena, Srimal; Sikka, Pavan; Hernandez, Emili. Plant Disease Detection using Hyperspectral Imaging. In: 2017 International Conference on Digital Image Computing: Techniques and Applications (DICTA); 29 Nov-1 Dec 2017; Sydney, NSW, Australia. IEEE; 2017. 1-8.

*This project (VG15024) has been funded by Hort Innovation, using the Hort Innovation vegetable levy research and development levy and contributions from the Australian Government, co-investment from the Commonwealth Scientific and Industrial Organisation (CSIRO), the Queensland Department of Agriculture and Fisheries (QDAF). Hort Innovation is the grower-owned, not-for-profit research and development corporation for Australian horticulture.

Subscribe to our News via Email

Enter your email address to subscribe and receive notifications of new posts by email.