Ανάπτυξη μεθοδολογίας αποθήκευσης δορυφορικών εικόνων υψηλής ανάλυσης σε βάση δεδομένων για ταχεία ανάκτηση από εξυπηρετητή διανομής.

Σκουλαρίκης, Δημήτριος; Skoularikis, Dimitrios

dc.contributor.author	Σκουλαρίκης, Δημήτριος	el
dc.contributor.author	Skoularikis, Dimitrios	en
dc.date.accessioned	2016-10-19T07:50:58Z
dc.date.available	2016-10-19T07:50:58Z
dc.date.issued	2016-10-19
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/43839
dc.identifier.uri	http://dx.doi.org/10.26240/heal.ntua.12728
dc.rights	Default License
dc.subject	Δορυφορική εικόνα	el
dc.subject	Αλγόριθμος	el
dc.subject	Ομαδοποίηση	el
dc.subject	Γενικό χαρακτηριστικό	el
dc.subject	Τοπικό χαρακτηριστικό	el
dc.subject	Satellite image	en
dc.subject	Algorithm	en
dc.subject	Clustering	en
dc.subject	Global feature	en
dc.subject	Local feature	en
dc.title	Ανάπτυξη μεθοδολογίας αποθήκευσης δορυφορικών εικόνων υψηλής ανάλυσης σε βάση δεδομένων για ταχεία ανάκτηση από εξυπηρετητή διανομής.	el
heal.type	bachelorThesis
heal.secondaryTitle	-	el
heal.classification	Γεωπληροφορική	el
heal.classification	Τηλεπισκόπηση	el
heal.language	el
heal.access	free
heal.recordProvider	ntua	el
heal.publicationDate	2016-09-12
heal.abstract	Η διπλωματική πραγματεύεται την εξαγωγή χαρακτηριστικών (features) από δορυφορικές εικόνες υψηλής ανάλυσης και την αποθήκευση αυτών ως μια βάση δεδομένων. Ακολούθως, τα εξαχθέντα features ομαδοποιούνται (clustering) με διαφορετικές τεχνικές ομαδοποίησης με σκοπό να βρεθεί και να προταθεί μια μεθοδολογία ταχείας ανάκτησης δορυφορικών εικόνων από εξυπηρετητή διανομής.	el
heal.abstract	This thesis deals with the methodology and the theoretical background necessary to achieve rapid recovery of a sufficiently large number of satellite images that are stored in a database. It is worth noting that the use and application of various algorithms is the tool that will help in this direction. The purpose of this thesis is to propose a methodology that through a fairly large volume of satellite imagery and the implementation of an algorithm the user can at a relative speed (via the distribution server) to determine the area from where the satellite image. Specifically, the first chapter presents some introductory concepts on the satellite images and nature of electromagnetic wave which is the root cause of all produced images. Moreover, we refer to the importance of photo-interpretation being done as part of an image evaluation (satellite image or aerial). In the second chapter we analyze the architecture of an image retrieval based on the content, which differs from that based on the text. Furthermore, reference is made to the semantic gap created due to the inability of low-level features to describe high-level concepts. In the third chapter we follow the concepts and deepen the internal characteristics (features) of satellite images. These characteristics are those that affect the grouping to be done and will ultimately determine the classes that will integrate our imagery. Within this section are presented satellite images with their histograms. The fourth chapter refers to the clustering technology cluster as the algorithms included in this category. In particular, we make an analysis of the k-means, fuzzy c-means, spectral clustering and DBSCAN algorithms. At the same time, there is a brief description of these algorithms accompanied by their advantages and disadvantages, whereas is also been done a small compare among those three (fuzzy c-means, spectral clustering, DBCSAN) and the most popular and widely known k-means algorithm. The fifth chapter contains the applied part of this thesis. Moreover, we apply two of the aforementioned algorithms in a series of samples of the satellite images of our database. In this way, k-means and fuzzy c-means that are programmed in Matlab code are being applied in a series of features extracted from our satellite images. This procedure in being done in order to recommend the optimal method of recovery of such images from a distribution server. Finally, the sixth chapter draws some useful conclusions from the implementation of the aforementioned algorithms, and analyzes the reasons and causes of the results obtained.	en
heal.advisorName	Δουλάμης, Νικόλαος	el
heal.committeeMemberName	Καραθανάση, Βασιλεία	el
heal.committeeMemberName	Δουλάμης, Αναστάσιος	el
heal.academicPublisher	Εθνικό Μετσόβιο Πολυτεχνείο. Σχολή Αγρονόμων και Τοπογράφων Μηχανικών. Τομέας Τοπογραφίας. Εργαστήριο Τηλεπισκόπησης	el
heal.academicPublisherID	ntua
heal.numberOfPages	125 σ.	el
heal.fullTextAvailability	true