Paper Title
A Modified RBF Network With Optimal Clustering For Face Identification And Localization

Abstract
Automated face identification is a challenging problem which has received much attention during recent years due to its many applications in different fields. This paper describes a robust and efficient method for face identification using Radial Basis Function Network (RBBFN) with Optimal Clustering Algorithm (OCA) for training units and Back Propagation (BP) Learning for classification. OCA clusters the pre processed training patterns which are taken as input of the RBFN. Then a general approach, which determines the initial structure and parameter are presented. Then BP Learning Algorithm is presented which classifies the input person angles into person. The present method successfully recognizes faces with different angles and expressions. The method enjoys the general advantages of RBFN with OCA and BP Learning. The identification of different faces is fast. The present method is effective, efficient and the identification rate is moderate.