Click on “Download PDF” for the PDF version or on the title for the HTML version.

If you are not an ASABE member or if your employer has not arranged for access to the full-text, Click here for options.

Multispectral images were collected by an unmanned aerial vehicle over a commercial coffee plantation during the 2002 harvest season. Selected scenes were georegistered to a base map and a mosaic of the study area was created. Image segmentation was performed to identify and mask soil, shadow, and cloud pixels. The remaining pixels, representing sunlit canopy, were assumed to be a mixture of four components: green leaf, underripe fruit, ripe fruit, and overripe fruit. Field and laboratory instruments were used to measure the reflectance spectrum of each component. Based on these spectra, a ripeness index was developed for the airborne imagery that involved computing the per-pixel ratio of digital counts in spectral channels centered at 580 and 660 nm. Results were aggregated on a per-field basis. Mean ripeness index per field was significantly correlated with ground-based counts recorded by the grower, and to eventual harvest date. The results suggest that remote sensing methods may provide an alternative, more spatially comprehensive method for monitoring ripeness status and evaluating harvest readiness of this high-value agricultural commodity.