Identification of purple spot disease on asparagus crops across spatial and spectral scales
•Three levels of Earth observation were tested to identify asparagus purple spot disease.•In-situ spectroscopy was used to assess spectrum regions that most contribute to disease identification.•Within-field infection maps are produced using VHR satellite imagery.•Landsat imagery was used to deduce...
Gespeichert in:
Veröffentlicht in: | Computers and electronics in agriculture 2018-05, Vol.148, p.322-329 |
---|---|
Hauptverfasser: | , , , , , |
Format: | Artikel |
Sprache: | eng |
Schlagworte: | |
Online-Zugang: | Volltext |
Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
Zusammenfassung: | •Three levels of Earth observation were tested to identify asparagus purple spot disease.•In-situ spectroscopy was used to assess spectrum regions that most contribute to disease identification.•Within-field infection maps are produced using VHR satellite imagery.•Landsat imagery was used to deduce the infection levels of fields across a wide area.•These methodologies can be combined to produce a monitoring and application decision support system.
Purple spot caused by the fungus Stemphylium vesicarium is a disease that causes heavy damage in asparagus crops. It infects the leaves and stem and results in degradation of the quality in the harvested asparagus shoots, lowering their commercial value. Precautionary spraying with fungicides is practiced to prevent its appearance, posing a threat to consumers and the environment. The aim of this work was to evaluate the ability of visible and near-infrared field spectroscopy (UNISPEC-DC) to identify the appearance of the purple spot symptoms at the level of asparagus plants and the ability of very high resolution (Pleiades-1A) and wide area (Landsat 8) satellite images to map fungal distribution within asparagus fields and across the river plain, respectively. Field measurements were used as reference to select the optimum infection identification models. Results indicate that field spectroscopy and wide area remote sensing can be used to create sufficiently accurate quantification models of disease severity in asparagus plants (R2 = 0.83 and RMSE = 9.68%) or across asparagus fields (R2 = 0.85 and RMSE = 8.26%). The accuracy drops when using satellite data to quantify infection within field (R2 = 0.34 and RMSE = 13.13%). The contributing wavelengths, bands and indices can be utilized to provide timely information to farmers and agronomists in order to support precision plant protection applications. |
---|---|
ISSN: | 0168-1699 1872-7107 |
DOI: | 10.1016/j.compag.2018.03.035 |