Abstract
In some real-world personal identification applications requiring high security standards, multibiometrics can provide higher identification accuracy than single biometrics. Due to its good performance, palmprint identification has received a lot of attention among other biometrics technologies. Combining left and right palmprints for multibiometrics is easy to implement and can produce better results. However, previous studies did not explore this issue in depth. An innovative framework for performing multibiometrics by combining left and right palmprint images is presented in this paper. This framework integrated three kinds of scores generated from the left and right palmprint images to perform matching score-level fusion. These first two scores were generated from left and right palmprint images, respectively, and can be obtained using any palmprint identification method. However, the third score was obtained using a specialized algorithm. Due to the careful consideration of the left and right palmprint images, the proposed algorithm can effectively exploit the similarity between the left and right palmprints. Moreover, the proposed weighted fusion scheme allowed perfect identification performance to be obtained in comparison with previous palmprint identification methods.