An object-based image classification rule-set approach for olive orchard mapping in the Mediterranean
Melemenis, A.
Olea europea L., commonly known as olive tree, has been of major economic, cultural and ecological importance in the Mediterranean basin for millennia. Recently olive oil has been recognized for its nutritional and health advantages compared to other oils, so the increasing demand and globalization of the market has led to EU directives for constant management of olive orchards. This work has a twofold purpose: (1) the accurate mapping of olive orchards by developing an object-based binary classification (olives/not olives) rule set applied on Very High Resolution satellite imagery, and (2) the study of the rule set's transferability among the different imageries. A custom rule set was developed and applied on FormoSat-2, IKONOS and WorldView-2 imagery of the Akrotiri peninsula in Western Crete. An analysis of the overall and individual accuracy of the classification results for all three images reveals that the rule set developed for the object-based image analysis responds very well, producing accurate results. The transferability of the rule set is demonstrated through the results of the classification and the accuracy assessment.
Show more [+] Less [-]AGROVOC Keywords
Bibliographic information
This bibliographic record has been provided by Mediterranean Agronomic Institute of Chania