Early detection of plant virus infection using multispectral imaging and spatial\u2013spectral machine learning

Yao Peng,J Steen Hoyer,James Legg,Bruce Grieve,Linda Hanley-Bowdoin,Mary M Dallas,José T Ascencio-Ibáñez,Hujun Yin

doi:10.1038/s41598-022-06372-8

Abstract

Cassava brown streak disease (CBSD) is an emerging viral disease that can greatly reduce cassava productivity, while causing only mild aerial symptoms that develop late in infection. Early detection of CBSD enables better crop management and intervention. Current techniques require laboratory equipment and are labour intensive and often inaccurate. We have developed a handheld active multispectral imaging (A-MSI) device combined with machine learning for early detection of CBSD in real-time. The principal benefits of A-MSI over passive MSI and conventional camera systems are improved spectral signal-to-noise ratio and temporal repeatability. Information fusion techniques further combine spectral and spatial information to reliably identify features that distinguish healthy cassava from plants with CBSD as early as 28 days post inoculation on a susceptible and a tolerant cultivar. Application of the device has the potential to increase farmers’ access to healthy planting materials and reduce losses due to CBSD in Africa. It can also be adapted for sensing other biotic and abiotic stresses in real-world situations where plants are exposed to multiple pest, pathogen and environmental stresses.

Highlights

Cassava brown streak disease (CBSD) is an emerging viral disease that can greatly reduce cassava productivity, while causing only mild aerial symptoms that develop late in infection
This paper describes the application of a custom built active multispectral imaging (MSI) (A-MSI) device and a machine learning method that leverages both spectral and spatial information of the imagery data for early detection of CBSD
We have developed an active multispectral imaging (A-MSI) sensor system enhanced with machine learning as a screening platform for virus infection

Summary

Introduction

Cassava brown streak disease (CBSD) is an emerging viral disease that can greatly reduce cassava productivity, while causing only mild aerial symptoms that develop late in infection. We have developed a handheld active multispectral imaging (A-MSI) device combined with machine learning for early detection of CBSD in real-time. RGB imaging has been used to recognise visual symptoms of the disease[3], while plant nutritional conditions and metabolic or biotic changes due to disease may be reflected in certain spectral wavelengths beyond the RGB c hannels[3,4,5]. These subtle signs in vast amounts of spectral and spatial imaging data can be successfully detected using advanced machine learning techniques. Because the necrotic lesions or root rot are only discovered when cassava is harvested, farmers often do not know that their crop is infected with CBSD until harvest at 9–12 months after planting

Methods

Results

Conclusion