Διερεύνηση χαρτογράφησης Ποσειδωνίας με χρήση επιβλεπόμενης ταξινόμησης οπτικών δορυφορικών εικόνων

Abstract

Posidonia oceanica meadows are an important marine ecosystem that offers habitat to fish, other organisms, and shelter for threatened species. In this postgraduate dissertation the validity of Posidonia mapping was investigated through optical satellite images. In particular, mapped data deriving from the side scan sonar was used, through the portal of the Department of Lands and Surveys and the WMS service, as well as Landsat 8 satellite image for the region of Cyprus. Training data and areas of interest (ROI) were created, followed by supervised classification with the Spectral Angle Mapper, Mahalanobis Distance, Maximum Likelihood and Minimum Distance algorithms in ENVI 5.4 software. A sample of 1,000 random points was added to the study area and then a relative comparison was made between the algorithms. Finally, as there was no raw data to automate the comparison of algorithms with ground control points, a random selection of 30 points was made manually.