Use of UAV-borne multispectral data and vegetation indices for discriminating and mapping three indigenous vine varieties of the Greek Vineyard

In line with precision viticulture, in recent years new methods of vineyard management have been introduced, so as to optimize vine cultivation and production wine the highest quality. Following on methodologies developed for mapping other crop parameters, there is currently a growing research effort discrimination varieties, this information useful vineyard-scale management, local regional inventory planning purposes, application EU Directives, support certification high quality wines. This focuses developing methodology, based UAV-borne multispectral data, discriminating three varieties Attica, Greece, employing non-parametric classifiers, namely Random Forest (RF), Support Vector Machines (SVM) Spectral Angle Mapper (SAM), selected vegetation indices (VIs). The suggested methodology uses easy obtain process, cost-effective images relies mostly free open-source software. Study conclusions suggest that although used did not result accurate at pixel level, expressed by overall accuracy (OA) 61.6%, they nevertheless proved plot level. Therefore, it considered effective applications require such level mapping.