Back to search

Aristotle University Of Thessaloniki (AUTH)

Ανάπτυξη μεθόδων υπολογιστικής νοημοσύνης και σύντηξης δεδομένων με εφαρμογή στο πεδίο της μηχανικής βιοσυστημάτων

Abstract

dc:description

The main objective of the presented thesis is to develop methods of Computational Intelligence and Data Fusion with application in the Biosystems Engineering area for the non-destructive testing of agricultural products and crop condition monitoring. The developed methods are related to the combination of sensors with Artificial Intelligence architectures in Precision Agriculture. The developed Artificial Intelligence algorithms included Bio-inspired Hierarchical Neural Maps and Novelty Detection algorithms that are capable of detecting sudden changes in different conditions of Biosystems Engineering area. In this thesis, Hierarchical models of Self-Organizing Maps (SOMs) in two different applications of Biosystems Engineering area are presented which included wheat yield prediction by data fusion from ground sensors and satellite images and harvesting stage detection in lettuce plants. Moreover, it is already known that in many cases of Biosystems Engineering applications, the condition of crops is not already known through Data Mining approaches. For this reason, it is necessary to determine their position by intelligent sensors which are capable of detecting whether any features are already memorized by a learning structure or an entirely new situation. In this thesis, Novelty Detection algorithms were utilized for the detection of new situations such as: i) weed species detection based on spectral characteristics, ii) Septoria and water stress detection based on spectral data and yellow rust, and iii) nitrogen stress detection based on spectral characteristics. Finally, this thesis presents an intelligent multi-disease plant identification system which was developed by utilizing leaf images. This smart system used the Support Vector Machine entsemple compared to the Active Learning approximation for plant diseases identification. By this method it is concluded that the first approach is particularly based on Support Vector Machines and the Active Learning approach is capable of using any One Class Classifier. The proposed methods were able to overcome all the above problems by presenting an over 90 % high yield depending on the application. The presented methods have managed to solve several key problems that are related to Precision Agriculture and act as useful tools for future use in larger scale practices.

Degree

thesis:*
Grantor dc:publisher
Aristotle University Of Thessaloniki (AUTH)
Year dc:date
2016

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Pantazi, Xanthoula-Eirini
  • Πανταζή, Ξανθούλα-Ειρήνη

Subjects

dc:subject × 12

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/38883

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

Pantazi, Xanthoula-Eirini; Πανταζή, Ξανθούλα-Ειρήνη. Ανάπτυξη μεθόδων υπολογιστικής νοημοσύνης και σύντηξης δεδομένων με εφαρμογή στο πεδίο της μηχανικής βιοσυστημάτων. Aristotle University Of Thessaloniki (AUTH), 2016. http://hdl.handle.net/10442/hedi/38883