Detection and mapping of Ridolfia segetum Moris patches in sunflower (Helianthus annuus L.) crop using remote sensing techniques
In: Precision Agriculture ‘05Search for other papers by J. M. Peña-Barragán in
Current site
Google Scholar
PubMed
Search for other papers by F. López-Granados in
Current site
Google Scholar
PubMed
Search for other papers by M. Jurado-Expósito in
Current site
Google Scholar
PubMed
Search for other papers by L. García-Torres in
Current site
Google Scholar
PubMed
Purchase instant access (PDF download and unlimited online access):
Ridolfia segetum Moris is an umbelliferous weed, which is very frequent and persistent in most sunflower crops in Spain. The hyperspectral signatures of bare soil and sunflower and Ridolfia segetum at different phenological stages (from emergency to senescence) were collected at three dates (mid-May, mid-June and mid-July 2003) using a handheld field spectroradiometer and, lately, were analysed to select the wavelength, bands and vegetation indices for hyperspectral and multispectral discrimination within and between phenological stages of sunflower and R. segetum. Simultaneously to the field measurements, conventional-color and color-infrared aerial photographs of two fields located in Cordoba province were captured, and remote sensing techniques were applied to map Ridolfia segetum patches in sunflower. Two methods of classification (Spectral Angle Mapper and Class Separation) were checked and the confusion matrix was used to determine the accuracy of the weed maps. In the multispectral study, the highest different was obtained in mid-June. In mid-July, the ANVI index [(NIR-B)/(NIR+B)] showed statistical differences between bare soil, R. segetum and all corresponding sunflower phenological stages, but these results were not completely similar in the aerial photograph analysis. The Spectral Angle Mapper classification of the mid-June photographs permitted the discrimination of R. segetum patches in sunflower with the highest accuracy (over 90 %).