Generating Cancelable Multispectral Palmprint templates based on Cartesian Transformation

Cancelable biometric (CB) has been principally suggested to solve some concerns associated to security and privacy of stored templates in traditional biometric systems. Its basic idea is to design a transformation function that starts from the original template until reaching a pseudo-identities by using an infinite number of secret keys of the same user. The requirements that must be met in an optimal cancelable algorithm are : cancelability, irreversibility, diversity and unlinkablility . In this paper, we propose a cancelable palmprint templates by using Cartesian transformation(CT) of minutiae points. The proposed scheme’s validity is confirmed through verification using the PolyU multispectral palmprint database. The experimental results and security analysis confirms the efficiency of CT to secure palmprint templates.