Face recognition based on face partitions using Common Vector Approach

ERGİN S., GÜLMEZOĞLU M. B.

3rd IEEE International Symposium on Control, Communications and Signal Processing (ISCCSP 2008), St Julians, Malta, 12 - 14 Mart 2008, ss.624-628 identifier identifier

  • Yayın Türü: Bildiri / Tam Metin Bildiri
  • Cilt numarası:
  • Doi Numarası: 10.1109/isccsp.2008.4537300
  • Basıldığı Şehir: St Julians
  • Basıldığı Ülke: Malta
  • Sayfa Sayıları: ss.624-628
  • Eskişehir Osmangazi Üniversitesi Adresli: Evet

Özet

In this study, the four different partitions (right eye, left eye, nose and mouth) that include more important information than other parts in a face are cropped in order to overcome the large sample size problem for face recognition. The feature vectors of images of each individual are constructed from these four partitions and applied to Common Vector Approach (CVA) for the recognition purpose. The reasonable results imply that the feature vectors obtained from four partitions of faces can be efficiently used instead of the feature vectors constructed from whole face.