Improved classification accuracy of powdery mildew infection levels of wine grapes by spatial-spectral analysis of hyperspectral images

Abstract

Background: Hyperspectral imaging is an emerging means of assessing plant vitality, stress parameters, nutrition status, and diseases. Extraction of target values from the high-dimensional datasets either relies on pixel-wise processing of the full spectral information, appropriate selection of individual bands, or calculation of spectral indices. Limitations of such approaches are reduced classification accuracy, reduced robustness due to spatial variation of the spectral information across the surface of the objects measured as well as a loss of information intrinsic to band selection and use of spectral indices. In this paper we present an improved spatial-spectral segmentation approach for the analysis of hyperspectral imaging data and its application for the prediction of powdery mildew infection levels (disease severity) of intact Chardonnay grape bunches shortly before veraison.Results: Instead of calculating texture features (spatial features) for the huge number of spectral bands independently, dimensionality reduction by means of Linear Discriminant Analysis (LDA) was applied first to derive a few descriptive image bands. Subsequent classification was based on modified Random Forest classifiers and selective extraction of texture parameters from the integral image representation of the image bands generated. Dimensionality reduction, integral images, and the selective feature extraction led to improved classification accuracies of up to 0.998 ± 0.003 for detached berries used as a reference sample (training dataset). Our approach was validated by predicting infection levels for a sample of 30 intact bunches. Classification accuracy improved with the number of decision trees of the Random Forest classifier. These results corresponded with qPCR results. An accuracy of 0.87 was achieved in classification of healthy, infected, and severely diseased bunches. However, discrimination between visually healthy and infected bunches proved to be challenging for a few samples, perhaps due to colonized berries or sparse mycelia hidden within the bunch or airborne conidia on the berries that were detected by qPCR. Conclusions:An advanced approach to hyperspectral image classification based on combined spatial and spectral image features, potentially applicable to many available hyperspectral sensor technologies, has been developed and validated to improve the detection of powdery mildew infection levels of Chardonnay grape bunches. The spatialspectral approach improved especially the detection of light infection levels compared with pixel-wise spectral data analysis. This approach is expected to improve the speed and accuracy of disease detection once the thresholds for fungal biomass detected by hyperspectral imaging are established; it can also facilitate monitoring in plant phenotyping of grapevine and additional crops.

Hyperspectral Imaging to Assess the Presence of Powdery Mildew (Erysiphe necator) in cv. Carignan Noir Grapevine Bunches

“…In a study more similar to the one presented here, Knauer et al [23] improved the classification accuracy of powdery mildew in grapevine bunches using a spatial-spectral segmentation approach. The dataset comprised 30 bunches of Chardonnay vines collected shortly before veraison.…”

“…Likewise, the identification of grapevine diseases by spectral imaging techniques has been an object of study [21]. Detection of powdery mildew and downy mildew in leaves by multispectral imaging was investigated by [5,22], while other authors focused on the use of hyperspectral imaging for the detection of downy mildew in leaves [10] and powdery mildew in bunches [23,24].…”

A new image-based tool for the high throughput phenotyping of pollen viability: evaluation of inter- and intra-cultivar diversity in grapevine

“…Accurate data collection, if combined with genetic information, can greatly accelerate progress in breeding for yield and quality traits of new and better adapted cultivars [ 23 ]. Recent works have reported the use of high-throughput technologies for the phenotyping of diverse grapevine traits, including plant phenology [ 36 ], crop yield components [ 36 – 38 ], grape quality [ 35 , 39 , 40 ] and fungal disease resistance [ 41 – 43 ]. In this context, the use of efficient and objective image-based systems for pollen viability estimation may provide an alternative solution for the time-consuming phenotyping of this trait.…”

‘Macrobot’–an automated segmentation-based system for powdery mildew disease quantification

“…Some methods are based on measuring the enzymatic activity of the infected tissue , chlorophyll fluorescence (Brugger, Kuska, and Mahlein 2018) PCR of fungal genes (Wessling and Panstruga 2012) but more commonly optical sensors and computer vision approaches are used. Hyperspectral imaging is using the information about the reflectance of the tissues in a wide range of wavelengths and may visualize the disease symptoms in relatively early stages (Knauer et al 2017, Thomas et al 2018. The multispectral imaging is done with only a few but usually highly informative wavelengths thus greatly reducing the cost of equipment and the amount of raw data.…”