Palm Print Recognition Using K-Means Clustering with Global Features

In these days there is a risk that others can access the same information anywhere and anytime. Currently passwords, personal identification cards are used for personal identification. Now a days Biometric based recognition is the most popular human recognition pattern. Biometrics measures individual’s unique or behavioural characteristics to authenticate personal identity. It provides more reliable and efficient means of identity verification. The physical dimension of hand known as palm geometry contains information that is capable of authenticating the identity of an individual. The goal of biometrics verification system consists in deciding whether two characteristics belong to same person or not. In this case image can be used for verification purposes. The term global states that whole image of the palm is considered for verification. Main work in this case is the pre-processing of image, then extracting the features, creating the data set, k-mean clustering algorithm and back propagation algorithm is used and performance is compared in both the cases. Keywords—Biometric verification, Hand geometry, Palm recognition, Neural Networks, Pre-processing, Feature