A Feasible 3D Ultrasound Palmprint Recognition System for Secure Access Control Applications

Biometric recognition systems based on 3D palmprint captured with optical technology have been widely investigated in the last decade; however, they can provide information about external skin surface only. This limit be overcome by Ultrasound, which allows gaining depth of palm lines and verify liveness sample, making very hard to fake. In this work, a feasible system ultrasound images is proposed. Unlike previous wet setups, coupling between probe human realized through gel pad, permits comfortable precise positioning hand user. Collected are processed generate 2D at various under-skin depths. features then extracted from these images, experimenting different procedures, merged define template that contains lines’ information. Recognition performances were evaluated performing verification identification experiments home-made database composed 423 samples 55 volunteers. An EER rate 0.36% an accuracy 100% obtained. The suitability proposed secure access control applications finally discussed.