http://amarillo.tamu.edu/files/2010/12/Using-digital-image-analysis-and-spectral-reflectance.pdf

Using digital image analysis and spectral reflectance data to quantify damage by greenbug (Hemitera: Aphididae) in winter wheat

Understanding the response dynamics of plants to biotic stress is essential to improve management practices and breeding strategies of crops and thus to proceed towards a more sustainable agriculture in the coming decades. In this context, hyperspectral imaging offers a particularly promising approach since it provides non-destructive measurements of plants correlated with internal structure and biochemical compounds. In this paper, we present a cascade of data mining techniques for fast and reliable data-driven sketching of complex hyperspectral dynamics in plant science and plant phenotyping. To achieve this, we build on top of a recent linear time matrix factorization technique, called Simplex Volume Maximization, in order to automatically discover archetypal hyperspectral signatures that are characteristic for particular diseases. The methods were applied on a data set of barley leaves (Hordeum vulgare) diseased with foliar plant pathogens Pyrenophora teres, Puccinia hordei and Blumeria graminis hordei. Towards more intuitive visualizations of plant disease dynamics, we use the archetypal signatures to create structured summaries that are inspired by metro maps, i.e. schematic diagrams of public transport networks. Metro maps of plant disease dynamics produced on several real-world data sets conform to plant physiological knowledge and explicitly illustrate the interaction between diseases and plants. Most importantly, they provide an abstract and interpretable view on plant disease progression.

https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0116902&type=printable

Metro maps of plant disease dynamics--automated mining of differences using hyperspectral images.

Determining when to dig peanut (Arachis hypogea L.) is complicated because of its indeterminate growth habit. Pod mesocarp color is often used an indicator of pod maturation. However, this process is time consuming and is usually based on a relatively small subsample of pods from peanut fields. Research was conducted during 2003-2005 to determine if reflectance of the peanut canopy, using multispectral imaging (350-2500 nm), could be used as an indicator of pod maturation. The cultivars VA 98R and NC-V 11 were planted beginning in early May through early June during each year with reflectance and the percentage of pods at optimum maturity (percentage of pods with brown or black mesocarp color) determined in mid-September. The highest yield observed for VA 98R across the 3 yr of the experiment was noted when peanut was planted in mid-May rather than early or late May or when planted in early June when peanut was dug based on optimum pod maturity using pod mesocarp color. Pod yield for the cultivar NC-V 11 did not differ when comparing planting dates. For cultivar VA 98R, Pearson's correlations were significant for all bandwidth categories except the normalized difference vegetation index (NDVI) when reflectance was compared with percentage of mature pods. Reflectance for NC-V 11 was not significant for any of the correlations even though significant differences in the percentage of mature pods were noted in mid September when comparing planting dates. These data suggest that canopy reflectance could potentially aid in predicting pod maturation, but more research is needed to determine feasibility of this approach.

Peanut Response to Planting Date and Potential of Canopy Reflectance as an Indicator of Pod Maturation

The sensor-based monitoring of diseases under controlled conditions establishes an objective tool that allows a better understanding of the pathogen-plant interactions in different situations. The purpose of our work was to implement the presymptomatic detection of powdery mildew on wheat leaves shortly after fungus inoculation by spectral and time-resolved laser-induced fluorescence spectroscopy. In the general scope of plant phenotyping, we hypothesized that it is possible to discriminate between wheat genotypes that are either resistant or susceptible to powdery mildew. According to our results, the presymptomatic detection of powdery mildew on wheat leaves was accomplished, irrespective of genotype, as early as one day after inoculation using the fluorescence amplitude ratio F451:F522. Similarly, the ratios F451:F522, F522:F687, and F522:F736 of the half-bandwidth are also appropriate parameters. Furthermore, in the spectral range between 410 nm and 620 nm, the mean lifetime was significantly longer in inoculated leaves than it was in control leaves. Finally, the short-term (10-12 hour) increase of the fluorescence mean lifetime at 530 nm and 560 nm following the inoculation suggests that the speed of the plant reaction might be associated to its resistance to the pathogen. Based on this information, we conclude that determinations of ultraviolet, laser-induced fluorescence intensity and lifetime are suitable approaches to presymptomatically detect powdery mildew on wheat leaves one day after inoculation.

Presymptomatic Detection of Powdery Mildew Infection in Winter Wheat Cultivars by Laser-Induced Fluorescence:

Ray blight disease, caused by Phoma ligulicola var. inoxydablis, is a serious threat to the Tasmanian pyrethrum industry. The management of this disease relies upon the strategic application of fungicides in early spring. A range of fungicides were assessed for their efficacy in controlling ray blight disease in Tasmanian pyrethrum fields, and the primary objective of this study was to increase fungicide options available to growers in different resistance groups. Fungicides were assessed under in vitro conditions, within five replicated-plot field trials over three seasons (2004 to 2006) and in single-plot trials over eight fields in 2005. In each of the field trials, regular assessments of disease intensity (defoliation severity and the incidence of stems with ray blight), stem height, and the number of flowers produced on each stem were made using stems as the primary sampling unit. Canopy reflectance at 830 nm and the Difference Vegetative Index, measured using a handheld multispectral radiometer, also were used to compare fungicide effects on green leaf area. The effect of fungicides on the dry weight of flowers, pyrethrin content within the flowers, flower maturity, and pyrethrin yield were determined. Under in vitro conditions, boscalid reduced both conidial germination and mycelial growth at concentrations of at least 0.16 μg/ml. In field trials 1 and 2 (in 2004), the premixed formulation of pyraclostrobin + boscalid (Pristine) increased pyrethrin yield by an average of 79% compared with nontreated plots over the two locations. Furthermore, in single-plot trials, pyraclostrobin + boscalid increased pyrethrin yield by 134 and 60% compared with the industry-recommended protocol (single application of azoxystrobin at 150 g a.i./ha [Amistar WG] and two additional applications of a tank mixture of difenoconazole at 125 g a.i./ha [Score] and chlorothalonil at 1,008 liters a.i./ha [Bravo 720] at 14- to 21-day intervals) and nontreated plots, respectively. In field trials 3 (in 2005) and 4 and 5 (in 2006), similar yield benefits also were produced by applying pyraclostrobin (Cabrio SC) or boscalid (Filan) alone or in combination with chlorothalonil (Bravo 720) at 1.4 liters of product per hectare, regardless of the rates of pyraclostrobin (250 and 125 g a.i./ha) and boscalid (500 and 250 g a.i./ha) used. These data were used to recommend the incorporation of boscalid to improve the fungicide-based management of ray blight disease. This decreases the number of applications of both strobilurin and triazole fungicides which have been used extensively for the management of ray blight and other diseases in Tasmanian pyrethrum fields and are prone to fungicide resistance development.

https://apsjournals.apsnet.org/doi/pdf/10.1094/PDIS-92-6-0887

Improving Fungicide-Based Management of Ray Blight Disease in Tasmanian Pyrethrum Fields

Summary

Reflectance from barley field trial plots was measured to determine whether it could be used as an accurate estimate of disease and yield response to fungicide, and whether this was consistent over a range of agronomic factors. Three trials were used to test the effects of cultivar, assessment date, fungicide and dose, different disease and fertiliser. Correlations with predominant disease and resulting yield loss were high, often better than visual disease or green leaf area measurements alone. Certain ratios of two reflectance wavelength measurements were more robust predictors than single wavelengths. Cultivar and development stage had a large influence on measurements and calibration for such morphology-based characters would be essential for practical deployment. Whilst weather conditions and solar angle restrict opportunities for usage of reflectance tools such as the‘Cropscan Radiometer', it may prove useful in crop loss assessments as an additional objective tool.

S J Wale

Papers

Relationship between canopy reflectance and yield loss due to disease in barley.