Masi, Giuseppe (2016) Image Segmentation in a Remote Sensing Perspective. [Tesi di dottorato]

[img]
Preview
Text
tesiMasi.pdf

Download (46MB) | Preview
[error in script] [error in script]
Item Type: Tesi di dottorato
Lingua: English
Title: Image Segmentation in a Remote Sensing Perspective
Creators:
CreatorsEmail
Masi, Giuseppegiuseppe.masi@unina.it
Date: 31 March 2016
Number of Pages: 138
Institution: Università degli Studi di Napoli Federico II
Department: Ingegneria Elettrica e delle Tecnologie dell'Informazione
Scuola di dottorato: Ingegneria dell'informazione
Dottorato: Ingegneria elettronica e delle telecomunicazioni
Ciclo di dottorato: 28
Coordinatore del Corso di dottorato:
nomeemail
Riccio, Danieledariccio@unina.it
Tutor:
nomeemail
Scarpa, GiuseppeUNSPECIFIED
Date: 31 March 2016
Number of Pages: 138
Uncontrolled Keywords: Image Segmentation, Remote Sensing, SAR, Multi-Resolution, Multiresolution, Multi-Spectral, Clustering
Settori scientifico-disciplinari del MIUR: Area 09 - Ingegneria industriale e dell'informazione > ING-INF/01 - Elettronica
Area 09 - Ingegneria industriale e dell'informazione > ING-INF/03 - Telecomunicazioni
Date Deposited: 12 Apr 2016 22:06
Last Modified: 31 Oct 2016 11:14
URI: http://www.fedoa.unina.it/id/eprint/11038

Abstract

Image segmentation is generally defined as the process of partitioning an image into suitable groups of pixels such that each region is homogeneous but the union of two adjacent regions is not, according to a homogeneity criterion that is application specific. In most automatic image processing tasks, efficient image segmentation is one of the most critical steps and, in general, no unique solution can be provided for all possible applications. My thesis is mainly focused on Remote Sensing (RS) images, a domain in which a growing attention has been devoted to image segmentation in the last decades, as a fundamental step for various application such as land cover/land use classification and change detection. In particular, several different aspects have been addressed, which span from the design of novel low-level image segmentation techniques to the de?nition of new application scenarios leveraging Object-based Image Analysis (OBIA). More specifically, this summary will cover the three main activities carried out during my PhD: first, the development of two segmentation techniques for object layer extraction from multi/hyper-spectral and multi-resolution images is presented, based on respectively morphological image analysis and graph clustering. Finally, a new paradigm for the interactive segmentation of Synthetic Aperture Radar (SAR) multi-temporal series is introduced.

Actions (login required)

View Item View Item