University of the Aegean
Διερεύνηση τεχνικών τηλεπισκόπησης και ανάπτυξη μεθοδολογίας για την ακριβή εκτίμηση της καύσιμης ύλης με βάση τη βλάστηση χρησιμοποιώντας δύο εποχικές δορυφορικές εικόνες υψηλής ανάλυσης: εφαρμογή σε δεδομένα μεγάλης έκτασης
Abstract
dc:descriptionIn the present PhD thesis, high resolution spatial satellite imagery was used to map vegetation in order to assign fuel characteristics to each vegetation class. High spatial resolution satellite imagery of two seasons from the new generation multispectral imaging system RapidEye were used for mapping vegetation in five study areas: Rhodes, Kastoria, Chalkidiki, West Attica and Messinia (≈10,000 km2). Initially, a spatially adaptive classification approach was designed. The classification was based on the segmentation of the study areas into individual spatial processing entities. Then, the satellite images were stratified into natural and man-made environment areas via a semi-automatic object-oriented classification approach. For the classification of the natural vegetation a "fuzzy" or "soft" classification was applied. Finally, a post-classification comparison using the Normalized Vegetation Index (NDVI) was performed to map bi –seasonal natural vegetation change. To evaluate the accuracy of the classifications, a random sampling scheme was selected. The overall accuracy ranges between 78.5% (Chalkidiki) and 89.5% (Kastoria) (88% Rodos, 87.4% Western Athens, 84.7 Messinia), while Cohen’s Kappa values calculated as equal to 0.86 for Rhodes, 0.88 for Kastoria, 0.74 for Chalkidiki, 0.85 for West Attica and 0.79 for Messinia. The results show that the combination of Remote Sensing and GIS techniques developed in this PhD thesis is an effective approach for mapping natural vegetation in large areas with complex and heterogeneous landscapes such as in Greece.
Degree
thesis:*- Grantor dc:publisher
- University of the Aegean
- Year dc:date
- 2017
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Retsilidou, Olga
- Ρετσιλίδου, Όλγα
Subjects
dc:subject × 20- Ταξινόμηση
- Βλάστηση
- Δορυφορική τηλεπισκόπηση
- Υψηλή χωρική ανάλυση
- Χαρτογράφηση καύσιμης ύλης
- Κανονικοποιημένος δείκτης βλάστησης
- Κατάτμηση
- Ασαφής ταξινόμηση
- Classification
- Vegetation
- High spatial resolution
- Fuel mapping
- RapidEye
- NDVI
- Segmentation
- Fuzzy or ‘soft’ classification
- Φυσικές Επιστήμες
- Γεωεπιστήμες και Επιστήμες Περιβάλλοντος
- Natural Sciences
- Earth and Related Environmental Sciences
Rights
dc:rights- Statement dc:rights
-
- BY
- Language dc:language
- gre
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/41190