Facial Expression Recognition Using Local Zernike Moments

Akkoca Gazioğlu B. S., Gokmen M.

21st Signal Processing and Communications Applications Conference (SIU), CYPRUS, 24 - 26 April 2013 identifier

  • Publication Type: Conference Paper / Full Text
  • Country: CYPRUS
  • Istanbul Technical University Affiliated: Yes


This study shows that Local Zernike Moments which is a novel study and works in face recognition efficiently can be used for facial expression recognition. Local Zernike Moments (LZM) method based on the calculation of moments at every pixel of the image unlike traditional Zernike Moments produces one moment value over whole image. Promising results are obtained by classifying the feature vectors with KNN (K Nearest Neighbor), belongs to the facial images.