Facial Expression Recognition under Noisy Environment Using Gabor Filters

1 Electronics Department, Faculty of Electrical Engineering and Information Technology, University of Oradea, Universitatii ,1, 410087, Oradea, Romania, e-mail: [email protected] Electronics and Communications Faculty, „Politehnica” University of Timisoara, Bd. V. Pârvan Nr. 2, 300223 Timişoara, e-mail: [email protected] 3 Dept. of Informatics, Aristotle University of Thessaloniki, GR – 541-24, Box 451, Thessaloniki, Greece, e-mail: [email protected] Abstract – Facial expression recognition is a major task concerning human-computer interaction issue. Plenty of techniques were proposed to recognize an expression either in still images or image sequences. However, most of them were applied for images recorded under controlled recording conditions. This paper aims at describing Gabor filters’ application to extract facial features required to classify facial expression when the images are disturbed by various noise levels. The experiments indicate a satisfactory performance for Gabor filters when compared to another state-of-the-art method named principal component analysis (PCA).