Back to search

National Technical University of Athens (NTUA)

Ανάπτυξη οπτικών μεθόδων και αλγορίθμων τρισδιάστατης ανακατασκευής βιολογικών επιφανειών και ανίχνευσης χαρακτηριστικών προσώπου με εφαρμογές στη βιοϊατρική

Abstract

dc:description

The scope of this dissertation is the development of innovative methods of Computer Vision that will tackle basic problems in Biomedicine. As part of the research, two computer vision systems were developed, the first for the purpose of three-dimensional reconstruction of biological surfaces and the second for the detection and recording of reflex movements of biometric features of the human face with applications in the study of the prepulse inhibition phenomenon. These are two basic problems of Computer Vision that find many applications in the Biomedical Sciences. The first part of the dissertation describes the development of a three-dimensional reconstruction algorithm, specialized in the reconstruction of biological surfaces. This algorithm aims at the reconstruction of biological tissues of Lambertian reflectivity and constant albedo, implementing an innovative technique that falls into the general category of Shape from Shading (SfS) methods, and is called Shape from Isophotes (SfI). SfS techniques are reconstruction methods that extract a three-dimensional shape through the analysis of spatial changes in brightness (light and shadow) of a 2D image. The proposed SfI technique uses the isophote curves extracted from the two-dimensional image to calculate the vertical vectors on the surface of the object. The new method was tested on digital and real biological surfaces with very satisfactory results. The second part of the dissertation describes the design and implementation of a high frame-rate optical recording system to remotely capture the movement of biometric features of the human face, with applications on the study of the of prepulse inhibition phenomenon. The study of prepulse inhibition of the acoustic startle reflex is a key area of medical research in the field of psychophysical disorders, playing an important role in the development and control of new types of antipsychotic therapy. The proposed method uses a high frame-rate optical capturing devise in conjunction with optical flow and facial feature detection algorithms to record the subject's reactions to auditory stimuli. In this way a significant reduction in the degree of invasiveness that characterizes other methods is achieved, fact that significantly reduces the patient's discomfort during the examination. The weighting of the new method was performed in parallel with a standard electromyography device, in order to compare the results of the two methods and to draw conclusions regarding the validity of the proposed method. The application of the proposed method on a sample of 32 subjects showed very positive results.

Degree

thesis:*
Grantor dc:publisher
National Technical University of Athens (NTUA)
Year dc:date
2020

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Balogiannis, Georgios
  • Μπαλογιάννης, Γεώργιος

Subjects

dc:subject × 24

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/48057

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

Balogiannis, Georgios; Μπαλογιάννης, Γεώργιος. Ανάπτυξη οπτικών μεθόδων και αλγορίθμων τρισδιάστατης ανακατασκευής βιολογικών επιφανειών και ανίχνευσης χαρακτηριστικών προσώπου με εφαρμογές στη βιοϊατρική. National Technical University of Athens (NTUA), 2020. http://hdl.handle.net/10442/hedi/48057