The best of both worlds: Applying secure sketches to cancelable biometrics

Cancelable biometrics and secure sketches have been introduced with the same purpose in mind: to protect the privacy of biometric templates while keeping the ability to match this protected data against a reference. The paradigm beyond cancelable biometrics is to perform an irreversible transformation over images and to make matching over transformed images. On one hand, a drawback of this technique is that for biometrics using a matching algorithm relying on some complex characteristics, such as the ones used for fingerprints, the irreversible transformation tends to break the underlying structure, thus degrading the performance accuracy. On the other hand, for secure sketches, matching is reduced to an error correction and we show here that applying secure sketch error correction to cancelable biometrics allows one to keep good matching performance. Moreover, the security’s advantages of both schemes adds up together.