| dc.description.abstract | This study was aimed at estimating area under coconut cultivation in Kaloleni Sub-County by
employing the use of GIS and remote sensing techniques. The viability of high resolution
satellite imagery (GeoEye) was tested to locate the distribution of the coconut trees in the study
area. After image pre-processing, the image was subjected to maximum likelihood classification
to produce a land cover map showing the distribution of coconut palm trees and other land-cover
types.
The classifier was trained and the accuracy of the results assessed using ground truth data of the
coconut palm trees and other land-cover types that were collected. Ground truth data, in the form
of field mappings, high resolution reference data and expert knowledge were used to gain further
insights and produce reliable results. Results of the supervised classification showed an overall
classification accuracy of 76.67%. A total of 303.82 hectares of coconut palms were detected
from the image analysis. | en_US |
