Back to search

Aristotle University Of Thessaloniki (AUTH)

Υπερφασματική ανάλυση δορυφορικών εικόνων για τη μελέτη, ποιοτικά και ποσοτικά, φυτικών παραμέτρων με τη χρήση GIS

Abstract

dc:description

Nowadays, the spaceborne hyperspectral data are a research challenge in various scientific fields and their assessment will contribute to the operational implementation. Thus taking into account the scientific interest for this type of data, the aim of the thesis was the investigation of Hyperion data, the first spaceborne hyperspectral sensor, in identification and classification of five agricultural crops (corn, cotton, rice, tobacco and tomato. More specifically, the objectives were the following: 1) the hyperspectral data dimensionality reduction and the determination of 'optimal' narrowbands that best characterizes the examined crops, 2) accuracy assessment of the implemented classifications algorithms and 3) assessment of the hyperspectral data discrimination power relative to the Landsat 5 TM data through the comparison of classifications accuracy. The study was conducted in the agricultural area between County borders of Kavala and Xanthi that comprises a part of the Nestos River and located between the settlement of Toxotes and the settlement of Keramoti. The data dimensionality reduction of Hyperion's image bands was achieved with the use of Stepwise Discriminant Analysis (SDA) and Minimum Noise Fraction Transformation (MNF) and algorithms such as Maximum Likelihood (ML), Mahalanobis Distance (MD) and Spectral Angle Mapper (SAM) were used in order to perform the image data classifications. The results lead to recommending 27 spectral bands that best characterizes the agricultural crops (80% data reduction) and Overall Accuracies of 91-92 % were attained with the use (ML) and (MD) and 75 % with (SAM). Finally the hyperspectral data provide an increased accuracy of 11% over six non thermal band of Landsat 5 TM when classifying agricultural crops. Conclusively, come up that narrowbands seems to be more effective in classification of agricultural crops in relation to Landsat 5 TM broadbands however further research is needed with the inclusion of more than five agricultural crops and at different phenological stages.

Degree

thesis:*
Grantor dc:publisher
Aristotle University Of Thessaloniki (AUTH)
Year dc:date
2008

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Ntouros, Konstantinos
  • Ντούρος, Κωνσταντίνος

Subjects

dc:subject × 16

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/24529

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Ntouros, Konstantinos; Ντούρος, Κωνσταντίνος. Υπερφασματική ανάλυση δορυφορικών εικόνων για τη μελέτη, ποιοτικά και ποσοτικά, φυτικών παραμέτρων με τη χρήση GIS. Aristotle University Of Thessaloniki (AUTH), 2008. http://hdl.handle.net/10442/hedi/24529