Ιδρυματικό Αποθετήριο [SANDBOX]
Πολυτεχνείο Κρήτης
EN  |  EL

Αναζήτηση

Πλοήγηση

Ο Χώρος μου

Spatiotemporal filtering of multi-temporal images: application on MODIS sea surface temperature (SST) imagery

Partsinevelos Panagiotis, Miliaresis George

Πλήρης Εγγραφή


URI: http://purl.tuc.gr/dl/dias/1A711500-C4AC-4AD9-AA2B-CF40EBB6B69B
Έτος 2009
Τύπος Πλήρης Δημοσίευση σε Συνέδριο
Άδεια Χρήσης
Λεπτομέρειες
Βιβλιογραφική Αναφορά P. Partsinevelos, G. Miliaresis. (2009). Spatiotemporal filtering of multi-temporal images: application on MODIS sea surface temperature (SST) imagery. Presented at 5th International Workshop on the Analysis of Multi-temporal Remote Sensing Images. [Online]. Available: file:///C:/Users/Δημήτρης/Downloads/2009_multitemp.pdf
Εμφανίζεται στις Συλλογές

Περίληψη

In this paper a series of spatiotemporal (ST) filters are devised in order to retrieve change information from multitemporalimagery depicting continuous-field data. Based on common spatial filters, derivative filters are extended toinclude 3-dimensional applicability, while new complex filters are designed to assist information retrieval undervarious application perspectives. In addition, ST filtering does not concentrate on a single pixel where possibleuncertainty resides but is applied upon a pixel group lying inside a defined parallelogram. Hence, weight,parallelogram size and filter shape selection lead to varying information extraction, including merely temporal,abrupt, gradual, directional and user defined spatiotemporal change. ST filtering of multi-temporal imagery resultsin a new multi-change dataset depicted as a 3-dimensional cloud of points classified in magnitude and/or type. Thisdataset is further examined to capture and visually convey an overall summarized change behavior. Thus, a selforganizing map algorithm is utilized, spreading along the change space and forming a 3-dimensional representative -signature polyline. To demonstrate the applicability of the proposed ST filters, monthly averaged sea surfacetemperature (SST) Modis images throughout a three year-period are processed. Temperature changes are classifiedaccording to their magnitude and type in an attempt to capture the seasonal variability, trends and possibleanomalies of SST in the Aegean region of Greece.

Διαθέσιμα αρχεία

Υπηρεσίες

Στατιστικά