An Efficient Visual Cryptography Approach for Intermodal Biometric Verification

Biometric authentication is one of the important ways of acknowledging an individual using their behavioral and physical characteristics. It can range from fingerprint to an iris or retina image, hand geometry, gait or signature, or some other physiological characteristics. The biometric authentication can be further sub-classified into unimodal and multimodal biometric authentication. The framework of biometrics, which utilizes only one suggestive option for verification is termed unimodal biometric. On the other hand, a lot of research work has been going on for the last two decades on multimodal biometric systems where more than one biometric detail of an individual is collected and used in order to authenticate a person. In other words, it can be said that in a multimodal biometrics system, more than two or sometimes at least two biometric sources of one individual (i.e., iris, dermatoglyphics) are considered and evaluated. In our paper, we have implemented an efficient way of fusing two biometric images (human fused iris image of both two eyes and fused thumb impression of two hands) into one image and converted the color image into a binary image, and used the (2,2) Visual Cryptography Scheme two efficiently create two out of two shares. As part of the reconstruction, we have overlapped the two generated shares to generate the fused image. The fused image was further processed to authenticate the genuine person. The proposed approach reduces storage costs and uses simple steps to implement the project which is easy to understand by non-technical people as well.