Authentication for ID cards based on colour visual cryptography and facial recognition

Abstract

Modern identification cards can be used for a myriad of applications such as electronic passports, ATM cards or payment cards for public transportation. Despite their ease of use, user authentication is an important factor that must be taken into consideration. In addition to the use of passwords, biometric data such as fingerprints or iris images can also be included as part of a multi-factor authentication system. However, these methods require secure storage of the biometric template and active participation from the user. In this paper, we propose a new method of authentication for identification cards based on colour visual cryptography and facial recognition. A colour image of the user will be encrypted using visual cryptography and split into two share images, one of which will be stored on the database and the other will be stored in the card. When the card is placed onto a sensor, the two shares will overlap to uncover the original image. The recovered image is then fed into a trained facial recognition algorithm to verify the user's identity. The proposed method is evaluated based on recognition rate and runtime. Experimental results indicate the feasibility of the proposed method for practical application and can be used as a starting point for future work in the area.