Skip to main content

01.03.2018 | Image & Signal Processing | Ausgabe 3/2018

Journal of Medical Systems 3/2018

Image-based Analysis of Emotional Facial Expressions in Full Face Transplants

Journal of Medical Systems > Ausgabe 3/2018
Merve Bedeloglu, Çagdas Topcu, Arzu Akgul, Ela Naz Döger, Refik Sever, Ozlenen Ozkan, Omer Ozkan, Hilmi Uysal, Ovunc Polat, Omer Halil Çolak
Wichtige Hinweise
This article is part of the Topical Collection on Image & Signal Processing


In this study, it is aimed to determine the degree of the development in emotional expression of full face transplant patients from photographs. Hence, a rehabilitation process can be planned according to the determination of degrees as a later work. As envisaged, in full face transplant cases, the determination of expressions can be confused or cannot be achieved as the healthy control group. In order to perform image-based analysis, a control group consist of 9 healthy males and 2 full-face transplant patients participated in the study. Appearance-based Gabor Wavelet Transform (GWT) and Local Binary Pattern (LBP) methods are adopted for recognizing neutral and 6 emotional expressions which consist of angry, scared, happy, hate, confused and sad. Feature extraction was carried out by using both methods and combination of these methods serially. In the performed expressions, the extracted features of the most distinct zones in the facial area where the eye and mouth region, have been used to classify the emotions. Also, the combination of these region features has been used to improve classifier performance. Control subjects and transplant patients’ ability to perform emotional expressions have been determined with K-nearest neighbor (KNN) classifier with region-specific and method-specific decision stages. The results have been compared with healthy group. It has been observed that transplant patients don’t reflect some emotional expressions. Also, there were confusions among expressions.

Bitte loggen Sie sich ein, um Zugang zu diesem Inhalt zu erhalten

e.Med Interdisziplinär

Mit e.Med Interdisziplinär erhalten Sie Zugang zu allen CME-Fortbildungen und Fachzeitschriften auf Zusätzlich können Sie eine Zeitschrift Ihrer Wahl in gedruckter Form beziehen – ohne Aufpreis.

Über diesen Artikel

Weitere Artikel der Ausgabe 3/2018

Journal of Medical Systems 3/2018 Zur Ausgabe