Advanced mathematical algorithms to characterize olive varieties through morphological parameters

The morphological analysis of olive fruits, leaves and endocarps may represent an efficient tool for the characterization and discrimination of varieties and the establishment of phenotypic relationships among them. In recent years, much attention has been focused on the application of DNA molecular markers due to their high capacity to efficiently and reliably discriminate cultivars. Here, we present a semi-automatic method of detecting various morphological parameters based on image analysis tools. A number of morphological parameters have been used to characterize olive germplasm collections from different countries. To date, for the morphological analysis of olives, old fashioned manual techniques (e.g. using screw gauge, gridded paper, etc.) have been used, or other unsuitable methods/softwares for handling the problem, due to the imposition of some prerequisites (e.g. color of the images background, position of the object, etc.). The novelty and the significance of our methodology lies in the fact that it is the first integrated methodology that provides automated morphological analysis of fruits, leaves and endocarps, without pre-processing manual tasks (cutting the fruits, etc.) or imposing prerequisites, after a manual binarised of the initial image. The underlying methodology is based on robust mathematical descriptors that can provide more accurate,rapid and consistent results regarding the shape description. In addition, the objective of this work is to serve as a useful further step in the development of computer-based techniques which can describe the whole morphology of crop species.