Aiming at the problem that exist technology in feature extraction

feature selection

pattern classification and identification

the obtained eigenvalue is not the optimal solution

and the rate of false recognition in identification is high

biometric identification

A new algorithm for feature selection of palm impedance spectrum is first presented by building an effective ellipse model（ EM） for the spectrum data of palm electrical resistance

which is based on the palm impedance spectrum biological data

and then fuse the immune clone technology and the least square method. The experimental results show that

the proposed algorithm not only can fully exploit the classification characteristics that the immune clone algorithm can be converged to the global optimal solution

but also can effectively achieve the feature extraction and feature selection for EM feature

finally

the rate of correct recognition is increased to 95 %. So the proposed algorithm has important theoretical and application value in information security

e-commerce and other fields.