A modify correlation filters for face verification

Abstract EN

Face verification is an important tool for authentication of an individual and it can be of significant value in security and e-commerce applications.In this paper, a new method for a minimum average correlation energy filter using a reduced number of training imagesto provide high discrimination capability and a sharp detected correlation peaks with shift invariance and variation tolerance characteristics with a high performance rate reached to 95% .The performance of the modified filter was evaluated by calculating the peak to side lobe ratio, the false acceptance ratio, the false rejection ratio and the equal error rate.

The computational experiments were done on a (Mat lab 2008) on the FACE 94 database