Contactless Authentication Device using Palm Vein and Palm Print Fusion Biometric Technology for Post Covid World

This paper proposes a secure, contactless authentication and biometric identification system incorporating palm vein and palm print image fusion. It acts as a security device without the need to physically touch it. In a world still recuperating from the aftermath of the COVID-19 pandemic, our no-contact device would find particular use in office spaces, educational institutes, etc. The system comprises near-infrared (NIR), light-emitting diode (LED) arrays, a low-cost complementary metal-oxide-semiconductor (CMOS) visual sensor, a ‘Raspberry Pi SBC’, and other components. In real-time, the palm vein and palm-print images are acquired at the same time with real-time Image Quality Assessment and Correction (IQAC). Vein and palm print features of interest are then made prominent using certain preprocessing algorithms. To reduce image storage requirements, we implement a wavelet-based fusion strategy on the images. Finally, using a deep scattering convolutional network, features are extracted. Although at present, a system uses state vector machines for training and recognition, this article advances the state-of-the-art method by introducing reduced vector machines to biometric authentication applications. Characteristics of some other biometric identification systems are also discussed to highlight features. The operational requirements of our proposed prototype are significantly lesser than fusion strategies employed by earlier technologies. It also proves to be more economical when implemented on a large scale, thus increasing the commercial viability of our system.