Back to search

Aristotle University Of Thessaloniki (AUTH)

Ταξινόμηση κάλυψης γης δορυφορικών εικόνων βασισμένη στα αντικείμενα με χρήση τεχνικών υπολογιστικής νοημοσύνης

Abstract

dc:description

This doctoral dissertation deals with the problem of land cover classification fromremote sensing images. A recently promising method is to consider the spatial informationexisting in the image, through the so called object-based approach, i.e. by applyingan image segmentation algorithm. In this context, a novel segmentation algorithm,called GeneSIS, is proposed in this dissertation, which refines the classification resultobtained initially by a pixel-wise classifier. This is achieved by exploiting the capabilitiesof computational intelligence methods, such as fuzzy logic and genetic algorithms.The GeneSIS algorithm creates a unique segmentation map, instead of a hierarchy, resolvingthe multi-scale problem in a single run. The proposed segmentation algorithmis embedded in four different classification schemes, each of which tries to address thesegmentation issue from a different point of view. The first scheme is applied on a pixelbasedmap, while the second one is applied on an already fine segmented map. Thethird methodology decomposes the overall process into a successive and independentextraction of unique classes and combines the obtained results at a subsequent stage.Finally, the fourth proposed scheme combines the properties of GeneSIS with the principlesof region growing algorithms, with the goal to boost the segmentation processwhile retaining high classification accuracies. The proposed schemes have been appliedto a diverse set of remote sensing images (agricultural,forest,urban) emanating fromdifferent sensors with variable spectral and spatial resolution.

Degree

thesis:*
Grantor dc:publisher
Aristotle University Of Thessaloniki (AUTH)
Year dc:date
2015

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Mylonas, Stylianos
  • Μυλωνάς, Στυλιανός

Subjects

dc:subject × 18

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/36816

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Mylonas, Stylianos; Μυλωνάς, Στυλιανός. Ταξινόμηση κάλυψης γης δορυφορικών εικόνων βασισμένη στα αντικείμενα με χρήση τεχνικών υπολογιστικής νοημοσύνης. Aristotle University Of Thessaloniki (AUTH), 2015. http://hdl.handle.net/10442/hedi/36816