Moshou_nutrient-pathogen-discrimination_PrecAgr2006 -...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
Simultaneous identifcation oF plant stresses and diseases in arable crops using proximal optical sensing and selF-organising maps D. Moshou Æ C. Bravo Æ S. Wahlen Æ J. West Æ A. McCartney Æ J. De Baerdemaeker Æ H. Ramon Published online: 20 April 2006 Ó Springer Science+Business Media, LLC 2006 Abstract The objective of this research was to detect plant stress caused by disease infestation and to discriminate this type of stress from nutrient deFciency stress in Feld conditions using spectral re±ectance information. Yellow Rust infected winter wheat plants were compared to nutrient stressed and healthy plants. In-Feld hyperspectral re±ectance images were taken with an imaging spectrograph. A normalisation method based on re±ectance and light intensity adjustments was applied. ²or achieving high performance stress identiFcation, Self-Organising Maps (SOMs) and Quadratic Discriminant Analysis (QDA) were introduced. Winter wheat infected with Yellow Rust was successfully recognised from nutrient stressed and healthy plants. Overall performance using Fve wavebands was more than 99%. Keywords Neural networks Æ Self-organising systems Æ Quadratic discriminant Æ Plant stresses Æ Plant diseases Æ Data mining Introduction Many pathogens cause discrete foci of disease within crops due to uneven survival or arrival of propagules or vectors, or the patchy distribution of favorable microclimates within the Feld (Waggoner & Aylor, 2000 ). In many polycyclic diseases, the dispersal of new propagules, predominantly around the original foci, intensiFes the development of patches of disease (McCartney & ²itt, 1998 ). However, it may take several pathogen generations for a small but visible focus of disease (0.5–2 m in diameter) to develop from a single lesion caused by a successful infection (Zadoks & Vandenbosch, 1994 ). Established D. Moshou ( & ) Æ C. Bravo Æ S. Wahlen Æ J. West Æ A. McCartney Æ J. De Baerdemaeker Æ H. Ramon Department of Biosystems, Division of Mechatronics, Biostatistics and Sensors (MeBioS), K.U. Leuven, Kasteelpark Arenberg 30, 3001 Heverlee, Belgium e-mail: [email protected] Precision Agric (2006) 7:149–164 DOI 10.1007/s11119-006-9002-0 123
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
infections within a crop become the predominant source of inoculum for new infections within the same crop (Rapilly, 1979 ) but in certain conditions, pathogens can spread rapidly over large distances (Brown & Hovmøller, 2002 ). Pesticides are commonly sprayed uniformly over the Feld, whilst most disease infes- tations occur in separate patches. Large ecological and Fnancial beneFts could be obtained if these patches could be treated in a site-speciFc way, setting healthy areas aside from spray treatment. This can be done by targeting pesticides only on those places in the Feld where they are needed based on disease mapping (West et al., 2003 ). Though airborne diseases tend to spread rapidly, targeting of pesticides based on disease mapping is a reasonable approach if it is combined with epidemic modelling to predict disease patch expansion (West et al., 2003 ). The use of epidemic modeling takes into account envi-
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 05/28/2010 for the course WE BIOL000000 taught by Professor Laurychaerle during the Spring '10 term at Ghent University.

Page1 / 16

Moshou_nutrient-pathogen-discrimination_PrecAgr2006 -...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online