Improving Human's Finger Knuckle Identification using High Order Zernike Moments

Identification systems based on biometrics have been conventional more than traditional identification systems in the last two decades. Many biometrics have been provided such as fingerprint, palm, iris, the vein of palm and veins of fingerprint and such. One of the challenges discussed in biometrics is physical damages. The biometric of fingers knuckles is one of the biometrics less exposed to the physical damages. Several methods have been suggested for identification with various weak points such as high mathematical complications and a very low rate of identification. The present study suggests a new method for identification which is based on Zernike Moment. Zernike moment extracts the features of the picture several times. What distinguishes this algorithm from its counterparts is that it has got high accuracy in demarcating similar pictures of different classifications. In addition toits logical calculating complications, this method was able to record a very appropriate rate of identification facing some challenges such as noise, rotation, and transition.