National Technical University of Athens (NTUA)
Tensor based training driven by semi-supervised graph based learning for efficient processing of high dimensional signals
Abstract
dc:descriptionThis dissertation proposes a framework combining tensor-based neural networks (Rank-R FNN via CP decomposition) with graph-based semi-supervised learning for hyperspectral image classification. The Rank-R FNN preserves spatial-spectral structure while reducing trainable parameters, and its embeddings are used to build similarity graphs for label propagation to unlabeled samples. Validated across four benchmark datasets in over 22,000 experimental runs, the framework achieves high classification accuracy even under limited labeled data conditions.
Degree
thesis:*- Grantor dc:publisher
- National Technical University of Athens (NTUA)
- Year dc:date
- 2026
Author and committee
dc:creator, dc:contributor.*- Authors dc:creator
-
- Georgoulas, Ioannis
- Γεωργούλας, Ιωάννης
Subjects
dc:subject × 20- Υπερφασματική απεικόνιση
- Τηλεπισκόπηση
- Μηχανική μάθηση
- Ταξινόμηση εικόνων
- Τεχνητά νευρωνικά δίκτυα
- Hyperspectral imaging
- Remote sensing
- Machine learning
- Image classification
- Artificial neural networks
- Άλλες Επιστήμες Μηχανικού και Τεχνολογίες
- Επιστήμες Μηχανικού και Τεχνολογία
- Τεχνολογία απεικόνισης και Φωτογραφική τεχνολογία
- Επιστήμη Ηλεκτρολόγου Μηχανικού, Ηλεκτρονικού Μηχανικού, Μηχανικού Η/Υ
- Υπολογιστές, Υλικό (hardware) και Αρχιτεκτονική
- Other Engineering and Technologies
- Engineering and Technology
- Imaging Science and Photographic Technology
- Electrical Engineering, Electronic Engineering, Information Engineering
- Computer science, Hardware and Architecture
Rights
- Language dc:language
- eng
Identifiers
dc:identifier.*- Identifier
- 10.12681/eadd/61775