In this chapter, faces are detected and facial features are located from video and still images. 'NRC-IIT Facial video database' is used as image sequences and 'face 94 color image database' is used for still images. Skin pixels and non-skin pixels are separated and skin region identification is done by RGB color space. From the extracted skin region, skin pixels are grouped to some meaningful groups to identify the face region. From the face region, facial features are located using segmentation technique. Orientation correction is done by using eyes. Parameters like inter eye distance, nose length, mouth position, and DCT coefficients are computed which is used for a RBF based neural network. © 2009 Springer Netherlands.
CITATION STYLE
Lekshmi, V. P., Sasikumar, M., & Vidyadharan, D. S. (2009). Face recognition and expression classification. In Lecture Notes in Electrical Engineering (Vol. 39 LNEE, pp. 669–679). https://doi.org/10.1007/978-90-481-2311-7_57
Mendeley helps you to discover research relevant for your work.