Publikationstyp: Konferenz: Poster
Art der Begutachtung: Keine Angabe
Titel: Budget remote phenotyping : potential and limitations of consumer-grade NIRGB and red edge cameras for early detection of plant leaf diseases
Autor/-in: Fahrentrapp, Johannes
Geilhausen, Martin
Laube, Patrick
Angaben zur Konferenz: 2nd General Meeting of COST Action FA 1306, Copenhagen, Denmark, 18-20 April 2016
Erscheinungsdatum: 18-Apr-2016
Sprache: Englisch
Schlagwörter: IUNR; Wein
Fachgebiet (DDC): 630: Landwirtschaft
Zusammenfassung: Plant leaf diseases can be very destructive and for many crops they lead to lower fruit quality and yield. To safe our harvests and income many preventive pesticides treatments are applied to manage diseases. Early detection could help to substantially reduce pesticide application to healthy crops allowing a site-specific treatment. Diseases in leaves can be detected and specified using hyperspectral images. Some approaches even allow the detection before eye-visible symptoms appear. These imagers are expensive, usually not very handy, and typically not constructed for field use. We analyzed series of images of Phytophthora infestans and Botrytis cinerea artificially inoculated tomato leaves within their first 48 hours of incubation. Images were taken with two different handheld, low cost Canon Power Shot cameras producing three band JPEGs (near-infra-red, green, blue, and red-edge, green, blue). At 24 hours past inoculation (hpi) first differences could be detected in single band reflectance between healthy (mock-inoculated) and pathogen-inoculated leaf disks. Significant differences could be found in reflectance range, mean and median values as well as in the index pigment specific simple ratio (PSSRb). Current findings will be presented.
Volltext Version: Publizierte Version
Lizenz (gemäss Verlagsvertrag): Lizenz gemäss Verlagsvertrag
Departement: Life Sciences und Facility Management
Organisationseinheit: Institut für Umwelt und Natürliche Ressourcen (IUNR)
Publiziert im Rahmen des ZHAW-Projekts: Multispektrale Bildgebung: Der Schlüssel zur Früherkennung von Blattkrankheiten
Enthalten in den Sammlungen:Publikationen Life Sciences und Facility Management

Dateien zu dieser Ressource:
Es gibt keine Dateien zu dieser Ressource.

Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt, soweit nicht anderweitig angezeigt.