Human facial expression recognition based on local MR8 filter bank

Facial expression represents intelligence, and facial expression recognition is always an important research field in computer vision and artificial intelligence for a long time. In this paper, we present a method of facial expression recognition based on local MR8 filter bank. At first, human face is detected by YIQ color space and morphologic features. Secondly, we find eyes and lip using morphologic features. Lastly, local features are measured by MR8 filter, and then facial expression is recognized. The method is computationally simply, easy to implement, which have quick detecting velocity and great recognition efficiency.