The Nearest feature line is a virtual line that connects two prototypes in a class (one person) while the midpoint feature is the midpoint between two prototypes in the same class. Thus, face treasury will be multiplied by conducting extrapolation or feature point interpolation on each feature line in the feature space. To perform this feature line based classification, as an initial representation of the image, the eigenface formation method will be used. In general, by forming an eigenface space, the dimensions that are significantly significant in the face image will be reduced so that only the important dimensions are left. The eigenface space formation cannot be separated from the use of Principal Component Analysis (PCA) as a dimensional reducing device.
Copyrights © 2018