A review of advanced machine learning methods for the detection of biotic stress in precision crop protection

Behmann, Jan; Mahlein, Anne-Katrin; Rumpf, Till; Römer, Christoph; Plümer, Lutz

doi:10.1007/s11119-014-9372-7

Cited by 305 publications

(181 citation statements)

References 77 publications

Supporting

Mentioning

179

Contrasting

Unclassified

Order By: Relevance

“…Another important contribution to precision agriculture is the development of mathematical tools that allow monitoring and classification of plants and fruits by the severity of the disease (Hahn, 2009), based on advanced statistical methods, as reviewed by Mulla (2013) and Behmann et al (2015). Some of these mathematical tools could be used as classifiers, identifying stressed plants, or monitoring the evolution of pests.…”

Section: Introductionmentioning

confidence: 99%

Multicolor Fluorescence Imaging as a Candidate for Disease Detection in Plant Phenotyping

et al. 2016

View full text Add to dashboard Cite

The negative impact of conventional farming on environment and human health make improvements on farming management mandatory. Imaging techniques are implemented in remote sensing for monitoring crop fields and plant phenotyping programs. The increasingly large size and complexity of the data obtained by these techniques, makes the implementation of powerful mathematical tools necessary in order to identify informative parameters and to apply them in precision agriculture. Multicolor fluorescence imaging is a useful approach for the study of plant defense responses to stress factors at bench scale. However, it has not been fully applied to plant phenotyping. This work evaluates the possible application of multicolor fluorescence imaging in combination with thermography for the particular case of zucchini plants affected by soft-rot, caused by Dickeya dadantii. Several statistical models -based on logistic regression analysis (LRA) and artificial neural networks (ANN)- were obtained for the experimental system zucchini-D. dadantii, which classify new samples as “healthy” or “infected.” The LRA worked best in identifying high dose-infiltrated leaves (in infiltrated and non-infiltrated areas) whereas ANN offered a higher accuracy at identifying low dose-infiltrated areas. To assess the applicability of these results to cucurbits in a more general way, these models were validated for melon infected by the same pathogen, achieving accurate predictions for the infiltrated areas. The values of accuracy achieved are comparable to those found in the literature for classifiers identifying other infections based on data obtained by different techniques. Thus, MCFI in combination with thermography prove useful at providing data at lab scale that can be analyzed by machine learning. This approach could be scaled up to be applied in plant phenotyping.

show abstract

Section: Introductionmentioning

confidence: 99%

Multicolor Fluorescence Imaging as a Candidate for Disease Detection in Plant Phenotyping

et al. 2016

View full text Add to dashboard Cite

show abstract

“…These more common methods are effective, although they are labor-intensive, time-consuming and often result in delayed diagnosis. In contrast, novel analysis based on imaging sensors can effectively detect early infections directly in the field [8,9]. In this context, imaging techniques are powerful non-destructive tools for precision agriculture, providing crucial information for decision-making and for determining the right timing for procedures to be applied [10].…”

Section: Imaging Techniques For Biotic Stress Detectionmentioning

confidence: 99%

“…In this context, imaging techniques are powerful non-destructive tools for precision agriculture, providing crucial information for decision-making and for determining the right timing for procedures to be applied [10]. These achievements are related both to the development of non-invasive, high-resolution optical sensors and of data analysis methods to deal with the large amount of data generated due to the high resolution, size and complexity rendered by these sensors [9]. Highly sophisticated and innovative methods of data analysis for complex plant-pathogen systems facilitate disease detection and expand the visual and/or molecular approaches for plant disease assessment [11].…”

Section: Imaging Techniques For Biotic Stress Detectionmentioning

confidence: 99%

“…Sensors measuring fluorescence, temperature or even reflectance can be useful tools for the identification, detection and quantification of plant diseases. Using several imaging techniques in parallel could allow disease detection in the absence of visible symptoms and therefore the early identification of emerging diseases in crops [9]. Chl-FI has successfully been used in biotic stress detection in the laboratory, as indicated by the vast amount of information on this topic in the literature.…”

Section: Future Prospects For Imaging Sensors In Plant Phenotyping Anmentioning

confidence: 99%

See 1 more Smart Citation

Picturing pathogen infection in plants

Barón

Pineda

Pérez-Bueno

2016

Zeitschrift Für Naturforschung C

View full text Add to dashboard Cite

Several imaging techniques have provided valuable tools to evaluate the impact of biotic stress on host plants. The use of these techniques enables the study of plant-pathogen interactions by analysing the spatial and temporal heterogeneity of foliar metabolism during pathogenesis. In this work we review the use of imaging techniques based on chlorophyll fluorescence, multicolour fluorescence and thermography for the study of virus, bacteria and fungi-infected plants. These studies have revealed the impact of pathogen challenge on photosynthetic performance, secondary metabolism, as well as leaf transpiration as a promising tool for field and greenhouse management of diseases. Images of standard chlorophyll fluorescence (Chl-F) parameters obtained during Chl-F induction kinetics related to photochemical processes and those involved in energy dissipation, could be good stress indicators to monitor pathogenesis. Changes on UV-induced blue (F440) and green fluorescence (F520) measured by multicolour fluorescence imaging in pathogen-challenged plants seem to be related with the up-regulation of the plant secondary metabolism and with an increase in phenolic compounds involved in plant defence, such as scopoletin, chlorogenic or ferulic acids. Thermal imaging visualizes the leaf transpiration map during pathogenesis and emphasizes the key role of stomata on innate plant immunity. Using several imaging techniques in parallel could allow obtaining disease signatures for a specific pathogen. These techniques have also turned out to be very useful for presymptomatic pathogen detection, and powerful non-destructive tools for precision agriculture. Their applicability at lab-scale, in the field by remote sensing, and in high-throughput plant phenotyping, makes them particularly useful. Thermal sensors are widely used in crop fields to detect early changes in leaf transpiration induced by both air-borne and soil-borne pathogens. The limitations of measuring photosynthesis by Chl-F at the canopy level are being solved, while the use of multispectral fluorescence imaging is very challenging due to the type of light excitation that is used.Keywords: biotic stress; chlorophyll fluorescence imaging; multicolor fluorescence imaging; plant pathogens; thermography. Dedication: This work is dedicated to the memory of Professor Peter Böger. He will always bring to my heart (M. B.) the best memories of my postdoc at Lake Constance.

show abstract

“…Water drought is one form that adversely affects crop productivity [1]. It occurs when the loss of water by transpiration exceeds that absorbed from the soil.…”

Section: Introductionmentioning

confidence: 99%

Suitable Image Features for Drought Stress Dection In Beans Using Raspberry Pi Imaging System

2017

IJIET

View full text Add to dashboard Cite

-Crop stresses are often detected late and this leads to massive crop yield loss. There exists a need for development of a real time system for monitoring crop growth progress with a view of early detection of diseases and other forms of crop stress. In this work, a simple raspberry pi imaging system is developed and used to capture images of beans subjected to drought stress. The images are used to extract features for classification of the stress. The features are then used to train classifiers for drought stress detection. Detection accuracy of between 90 to 100% has been achieved.

show abstract

A review of advanced machine learning methods for the detection of biotic stress in precision crop protection

Cited by 305 publications

References 77 publications

Multicolor Fluorescence Imaging as a Candidate for Disease Detection in Plant Phenotyping

Multicolor Fluorescence Imaging as a Candidate for Disease Detection in Plant Phenotyping

Picturing pathogen infection in plants

Suitable Image Features for Drought Stress Dection In Beans Using Raspberry Pi Imaging System

Contact Info

Product

Resources

About