|
In face recognition, one of the most important problems to tackle is a large amount of data and the redundancy ofinformation contained in facial images. There are numerous approaches attempting to reduce this redundancy. Oneof them is information aggregation based on the results of classi ers built on selected facial areas being the mostsalient regions from the point of view of classi cation both by humans and computers. In this study, we report ona series of experiments and o er a comprehensive comparison between various methods of aggregation of outputs ofthese classi ers based on essential facial features such as eyebrows, eyes, nose, and mouth areas. For each of them,we carry the recognition process utilizing the well-known Fisherfaces transformation. During the comparisons of thevectors representing the features of images (faces) after the transformations, we consider 16 similarity=dissimilaritymeasures for which we select the best aggregation operator. The set of operators to compare was selected on a basis ofthe comprehensive literature review regarding aggregation functions
|
