Jiangsu Tiera Biotechnology Co., Ltd , https://www.tierabio.com
Results Sixty barley varieties were studied by Specim V10E hyperspectral imaging 30 days after inoculation with powdery mildew. With high spatial resolution and stable measurement conditions, powdery mildew symptoms can be automatically quantified through the mechanical combination of Simplex Volume Maximization and Support Vector Machines. During the manual rating, once the first symptom is visible to the naked eye, it can be tested immediately. An accurate assessment of the disease severity of all cultivars per measurement day was achieved during the course of the experiment. In addition, a species of powdery mildew resistant to necrosis was detected.
Conclusion The hyperspectral phenotypic analysis system is a measurement system that combines the advantages of both on-site canopy level measurement systems (high throughput, automation, low manual workload) and laboratory-based blade levels (high spatial resolution, subject to Control the environment, the stability of time series measurements), so that accurate and objective disease severity assessment can be performed without the need for trained experts to assess – they can visually grade at an early stage and detect disease symptoms. Therefore, it is a promising and efficient research tool for plant resistance breeding.
Key words
Specim Hyperspectral Imaging, Phenotypic Analysis Platform, Greenhouse, High Throughput, Disease Grading, Simplex Volume Maximization, Support Vector Machine
The following are some of the graphic results of the study. 
Table 1 Manual rating of each barley variety disease during the experiment (2016; - means asymptomatic) 
The above content is extracted from research papers:
Stefan Thomas, Jan Behmann et al. Quantitative assessment of disease severity and rating of barley cultivars based on hyperspectral imaging in a non-invasive, automated phenotyping platform. Plant Methods (2018) 14:45.
Yi Ketai Ecological Technology Co., Ltd. provides comprehensive solutions for hyperspectral technology: 
High-spectrum high-throughput phenotypic analysis platform for visual assessment of the degree of disease in different varieties of barley
The background phenotype is the bottleneck in the development of new plant cultivars. This study introduces a new hyperspectral phenotypic analysis system that combines the high throughput of canopy measurement with the advantages of high spatial resolution and controllable measurement environment. Combine. In addition, the measured barley grows in large containers (called Mini-Plots), which allows plants to form field phenotypes in greenhouse experiments, independent of container size. Fig. 1 Kleinâ€Altendorf campus greenhouse phenotype Mini-Plot equipment, divided into greenhouse interior (a) and greenhouse outside (b), see the high-spectrum phenotypic analysis system (c). The track system of the Mini-Plot device, combined with a diffuse artificial light source and screen, enables automatic measurements in highly controlled environmental conditions. Specim V10E Hyperspectral Imaging (HSI) sensor combined with mirror-based scanner system for fast, high-resolution measurements of the entire Mini-Plot Fig. 2 Spectral characteristics and abundance map of healthy plant tissues and powdery mildew symptoms (average of 50 pixels/each). The spectral signature (left) represents the average reflectivity of the pixels on the spectral measurement area of ​​the sensor. The abundance map (right) shows the performance of the same pixel abundance average based on 25 prototypes selected during the conversion of the dataset using Simplex Volume Maximization. Prototypes that are highly correlated with healthy or symptomatic tissue are shown separately, each prototype is derived from the true spectral characteristics of the original hyperspectral dataset (the color of the prototype spectral features represents the color of the corresponding pixel, which is visible to the human eye), Pm stands for powdery mildew (powdery mildew) * Rating based on the list of official German varieties Fig. 3 Spectral characteristics of healthy and symptomatic tissues at different canopy levels (average of 30 pixels) Fig. 4 Disease severity of different barley cultivars during the experiment. Estimating disease severity based on pixel percentage, classified as mildew-containing symptoms after Simplex Volume Maximization (SiVM) and following Supervised Classification Support Vector Machines (SVM) Fig.5 Spatial distribution of high-susceptibility varieties Milford and Tocada (same period) after infection with powdery mildew during the experiment (16 days, 18 days, 22 days, 26 days), via pseudo RGB image and Support Vector Machines ( SVM) false color image classification (green for healthy tissue, red for powdery mildew symptoms, blue for background) Fig. 6 The classification of powdery mildew infection in healthy and necrotic lesions after 2 days of inoculation of barley resistant variety Irina. The false color picture shows the spatial distribution of necrotic lesions in the control and vaccinated plants in Mini-plots. In addition, the percentage of total pixels in the image is defined as the tissue of the allergic reaction is represented by the right histogram, and the vaccination group has a significant increase in the proportion of pixels defined as necrotic lesion expression.