A Hyperspectral Library of Foliar Diseases of Wheat

Abstract

This work established a hyperspectral library of important foliar diseases of wheat in time series to detect spectral changes from infection to symptom appearance induced by different pathogens. The data was generated under controlled conditions at the leaf-scale. The transition from healthy to diseased leaf tissue was assessed, spectral shifts were identified and used in combination with histological investigations to define developmental stages in pathogenesis for each disease. The spectral signatures of each plant disease that are indicative of a certain developmental stage during pathogenesis - defined as turning points - were combined into a spectral library. Different machine learning analysis methods were applied and compared to test the potential of this library for the detection and quantification of foliar diseases in hyperspectral images. All evaluated classifiers provided a high accuracy for the detection and identification for both the biotrophic fungi and the necrotrophic fungi of up to 99%. The potential of applying spectral analysis methods, in combination with a spectral library for the detection and identification of plant diseases is demonstrated. Further evaluation and development of these algorithms should contribute to a robust detection and identification system for plant diseases at different developmental stages and the promotion and development of site-specific management techniques of plant diseases under field conditions.

Applicability of Smart Tools in Vegetable Disease Diagnostics

“…Various computational algorithms and diagnostic models have been tested to improve visual evaluation and further identification [21,22]. The existence of precise databases of sufficiently adequate images has been [23] reported to be the basic prerequisite for enabling such a system to be functional. These precise tools for phytopathogen interception are still expensive and, in the case of vegetables, not sufficiently advanced.…”

Digital plant pathology: a foundation and guide to modern agriculture

“…To differentiate pathogen attack symptoms, resistance reactions, abiotic stress and spectral signatures of healthy leaves, each of these states had to be characterized in detail (Carter and Knapp 2001 ). Multi- and hyperspectral imaging is the preferable technique to study such interactions from the cell level to the canopy (Bohnenkamp et al 2019a , b , 2021 ).…”

“…Importantly, HSI comes with numerous advantages compared to classical visual monitoring or other analytical methods. It can be applied at different scales—from the cellular level for investigating plant tissue in combination with microscopes, over the individual plant scale in greenhouses or climate chambers, to the canopy scale in field applications with cameras mounted on unmanned aerial vehicles or airplanes (Bohnenkamp et al 2019a , b , 2021 ; Heim et al 2019a ). However, in all cases, the analysis of HSI data must be done with care as a great complexity results from a triangular relationship between sensor, pathogen, and environment (Fig.…”

“…Steps like de-noising, smoothing, calibration, image segmentation, and outlier removing must be added to transfer the image data to features that can be used as input for machine learning routines (Paulus and Mahlein 2020 , Behmann et al 2015 ). Literature shows not only the importance of this step but also the huge effort that is required for different sensors in greenhouses and in the field (Bohnenkamp et al 2021 , 2019a , b ; Thomas et al 2018b ). Hyperspectral measurements either use the raw reflectance signal as input or, for reduction of data complexity, vegetation indices like the NDVI or OSAVI (Bohnenkamp et al 2019a , b ; Rouse et al 1974 , Rondeaux 1996 ).…”

Plant Disease Sensing: Studying Plant-Pathogen Interactions at Scale

“…Untangling the processes that contribute the most to the spectral disease phenotype will one day lead to reliable previsual disease detection and differentiation from nonbiotic stress at scale. Case studies have shown that the presymptomatic disease phenotype can differ between infections caused by both different pathogen types ( 5 , 40 , 59 ) and different isolates ( 39 , 60 ) and can be strongly impacted by host genotype ( 38 , 61 , 62 ). How will these caveats affect regional and global disease monitoring?…”