Συντάχθηκε 11-07-2012 13:09
από Galateia Malandraki
Email συντάκτη: gmalandraki<στο>tuc.gr
Ενημερώθηκε:
-
Ιδιότητα: υπάλληλος ΑΡΜΗΧ.
ΠΟΛΥΤΕΧΝΕΙΟ ΚΡΗΤΗΣ
Τμήμα Ηλεκτρονικών Μηχανικών & Μηχανικών Υπολογιστών
ΠΑΡΟΥΣΙΑΣΗ ΔΙΠΛΩΜΑΤΙΚΗΣ ΕΡΓΑΣΙΑΣ
Παπαγρηγορίου Στυλιανός
με θέμα
“Αποδοτική Κατάτμηση και Κατηγοριοποίηση Hyper Spectral Κύβων.”
“Efficient Segmentation and Classification of Hyper Spectral Cubes.”
Τρίτη 17 Ιουλίου 2012 και 14:30
Αίθουσα 2041, Εξωτερικά του Κτιρίου Επιστημών, Πολυτεχνειούπολη
Εξεταστική Επιτροπή
Καθηγητής Γαροφαλάκης Μίνως (επιβλέπων)
Αν. Καθηγητής Μπάλας Κώστας
Επ. Καθηγητής Παπαευσταθίου Γιάννης
Abstract
Hyper Spectral Imaging is a powerful analytical tool, which has been used
in a wide area of applications, from Satellite Imaging to Biomedical Diagnosis.
By photographing the material to be examined and acquiring a Hyper Spectral
Cube (an "image" with information on more than the RGB spectrum) one is
able to extract information about the nature of the material, by studying its
optical impression on the Spectral Cube.
This technique offers a non-destructive and non-immediate way (one does
not have to extract part of the material and bring it to the lab) of examining
materials, something perfect for medical purposes.
In the hereby thesis the computational capabilities of spectral imaging methods
are examined and attempted to be improved, in order to provide real time
pixel classification. Specifically, a successful attempt is made to improve the
classification processing time of hyper spectral cubes acquired from a cervix
biopsy. The goal was to provide a four-class color map, with each class referring
to a specific cell condition.
Although this study was based on specific medical data, it is possible to be
generalized on any aspect of Hyper Spectral Imaging, and provides proof that
real-time Hyper Spectral Processing for classification purposes is feasible.
Συνημμένα:
-
Δ.Ε_ανακοινωση_παπαγρηγορίου.doc
Μεταφορτώσεις: 166,
Μέγεθος: 44 KB application/msword