Back to search

University of Patras

Αναγνώριση ψυχοσωματικής κατάστασης προσώπου με τη χρήση θερμικής υπέρυθρης ακτινοβολίας

Abstract

dc:description

In this PhD Thesis various algorithms are proposed and their discrimination capabilities are investigated for drunk persons identification using thermal infrared images of the persons’ faces. The original idea lies on the fact that the blood vessels’ network of the face will present increased activity when the person has consumed alcohol. This could lead to an automated system for discriminating intoxicated persons. Face recognition in the visible spectrum is strongly affected by the conditions of illumination. Acquiring images from faces in thermal infrared spectrum information related to the temperature of the face is obtained which mainly depends on the physiological condition of the person. The human face being in a mean temperature around 300 K, radiates according to the Wien Law as a perfect black body, with maximum at 10μm wavelength. Drunkenness is a challenging physiological condition to be investigated using infrared imagery. However, most of the publications in the literature refer only to automotive anti-drunk driving systems, which utilize electrical signals from the heart or brain. In this PhD Thesis thermal infrared images were employed by means of the thermal infrared camera Thermo Vision Micron/A10 (18mm, F/1.6). During the development of the PhD Thesis, 41 persons consumed alcohol in a controlled way in order to create our database (http://www.physics.upatras.gr/sober/). Seven different approaches are proposed in this thesis to be assessed for identifying intoxication by means of thermal infrared images. In the first these chapters, introductory material is given. In Chapter 4, a simple feature vector was formed for drunk person identification by simply taking 20 different points on the face of each person. Using the Fisher Linear Discriminant Analysis the features space is decreased from twenty to two dimensions, which correspond to the largest eigenvalues. It is apparent that the corresponding clusters for each person after alcohol consumption are moving towards the same direction in the feature space. Accordingly, the feature space can be separated into two regions "drunk" and "sober". In Chapter 5, the temperature differences which are presented on the face after alcohol consumption are discussed. From the experimental results, we observed that specific regions of the face, compared with other regions, increase their temperature after alcohol consumption e.g. forehead-nose, mouth-forehead. Specifically, according to this approach it is not necessary to have the thermal infrared image corresponding to the sober person. In Chapter 6, drunk person identification is examined by means of the activity of blood vessels on the face when the person is drunk. Thermal images are employed to assess the increased activity of the facial vessels which appear brighter when the person consumes alcohol. In Chapter 7, neural networks are tested on infrared images of faces for discriminating drunk persons. Neural networks of the same structure but of different sizes are employed to accommodate the training samples. It is proved in this work that a small neural structure trained on the forehead of a specific person (sober and drunk) presents very good generalization performance and discriminates between sober and drunk without having been trained on the specific person beforehand (90%). In Chapter 8, isothermal regions on the face of sober and intoxicated persons are extensively studied for drunk person identification. It is found that the isothermal regions on the face are changing with alcohol consumption. In Chapter 9, temperature distribution on the eyes of sober and drunk persons is studied by means of thermal infrared images. The temperature difference between the sclera and the iris is examined extensively. Both regions of the eye are of the same temperature for the sober person. For the intoxicated person the iris becomes darker, which means that the sclera temperature increases. Markov procedures are employed in Chapter 10 for the discrimination of drunk persons. In chapter 11, a bank of velocity filters is devised to be used for isolating a moving object with specific velocity (amplitude and direction) in a sequence of frames.

Degree

thesis:*
Grantor dc:publisher
University of Patras
Year dc:date
2014

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Koukiou, Georgia
  • Κουκίου, Γεωργία

Subjects

dc:subject × 24

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/39864

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

Koukiou, Georgia; Κουκίου, Γεωργία. Αναγνώριση ψυχοσωματικής κατάστασης προσώπου με τη χρήση θερμικής υπέρυθρης ακτινοβολίας. University of Patras, 2014. http://hdl.handle.net/10442/hedi/39864