Login | Request Account (DAF staff only)

Detecting myrtle rust (Austropuccinia psidii) on lemon myrtle trees using spectral signatures and machine learning

Share this record

Add to FacebookAdd to LinkedinAdd to XAdd to WechatAdd to Microsoft_teamsAdd to WhatsappAdd to Any

Export this record

View Altmetrics

Heim, R. H. J., Wright, I. J., Chang, H. ‐C., Carnegie, A. J., Pegg, G. S., Lancaster, E. K., Falster, D. S. and Oldeland, J. (2018) Detecting myrtle rust (Austropuccinia psidii) on lemon myrtle trees using spectral signatures and machine learning. Plant Pathology, 67 (5). pp. 1114-1121. ISSN 0032-0862

Full text not currently attached. Access may be available via the Publisher's website or OpenAccess link.

Article Link: https://doi.org/10.1111/ppa.12830

Publisher URL: https://doi.org/10.1111/ppa.12830

Abstract

Hundreds of species in one of Australia's dominant plant families, the Myrtaceae, are at risk from the invasive pathogenic fungus Austropuccinia psidii. Since its arrival in Australia in 2010, native plant communities have been severely affected, with highly susceptible species likely to become extinct from recurring infections. While severe impact on Australian native and plantation forestry has been predicted, the lemon myrtle industry is already under threat. Commercial cultivars of lemon myrtle (Backhousia citriodora) are highly susceptible to A. psidii. Detecting and monitoring disease outbreaks is currently only possible by eye, which is costly and subject to human bias. This study aims at developing a proof?of?concept for automated, non?biased classification of healthy (naïve), fungicide?treated and diseased lemon myrtle trees by means of their spectral reflectance signatures. From a lemon myrtle plantation, spectral signatures of fungicide?treated and untreated leaves were collected using a portable field spectrometer. A third class of spectra, from naïve lemon myrtle leaves that had not been exposed to A. psidii, was collected from a botanical garden. Reflectance spectra in their primary form and their first?order derivatives were used to train a random forest classifier resulting in an overall accuracy of 78% (kappa = 0.68) for primary spectra and 95% (kappa = 0.92) for first?order derivative?transformed spectra. Thus, an optical sensor?based discrimination, using spectral reflectance signatures of this as yet uninvestigated pathosystem, seems technically feasible. This study provides a foundation for the development of automated, sensor?based detection and monitoring systems for myrtle rust.

Item Type:Article
Business groups:Horticulture and Forestry Science
Keywords:derivative spectrometry field spectroscopy plant pathogens precision agriculture random forest rust fungi
Subjects:Plant culture > Tree crops
Plant pests and diseases > Plant pathology
Plant pests and diseases > Pest control and treatment of diseases. Plant protection
Technology > Technology (General) > Spectroscopy
Live Archive:17 Aug 2018 05:01
Last Modified:03 Sep 2021 16:51

Repository Staff Only: item control page