A New Clustering Algorithm for Face Classification

In This paper, we proposed new clustering\udalgorithm depend on other clustering algorithm ideas.\udThe proposed algorithm idea is based on getting distance\udmatrix, then the exclusion of the matrix points which will\udbe clustered by saving the location (row, column) of these\udpoints and determine the minimum distance of these\udpoints which will be belongs the group (class) and keep\udthe other points which are not clustering yet. The propose\udalgorithm is applied to image data base of the human face\udwith different environment (direction, angles... etc.).\udThese data are collected from different resource (ORL\udsite and real images collected from random sample of\udThi_Qar city population in lraq). Our algorithm has been\udimplemented on three types of distance to calculate the\udminimum distance between points (Euclidean,\udCorrelation and Minkowski distance) .The efficiency\udratio of proposed algorithm has varied according to the\uddata base and threshold, the efficiency of our algorithm is\udexceeded (96%). Matlab (2014) has been used in this\udwork.\u