3D Ultrasound palm vein pattern for biometric recognition Antonio Iula S.I. – University of Basilicata Viale dell’Ateneo Lucano 10, 85100 Potenza, ITALY Antonio.iula@unibas.it Alessandro Savoia, Giosuè Caliano, DIE – University Roma Tre, Roma, ITALY Abstract— In this work, an ultrasound technique for achieving 3D palm vein patterns for biometric recognition purposes is proposed and experimentally tested. Commercial ultrasound imaging machine and linear arrays have been employed. Both imaging and power Doppler analysis have been performed to detect veins. The probe is moved in the directional orthogonal to the array by an automated scanning system and at each step a a 2D frame is captured and stored to form a 3D matrix. The data from the 3D matrix are elaborated for achieving 3D ultrasonic vein patterns. The proposed technique has been applied to acquire hand vein patterns of about a dozen of distinct volunteers, which are opportunely rendered and discussed. Keywords-Biometrics; Vein pattern; Acoustic Imaging; Ultrasonic Array. I. INTRODUCTION Biometrics can be defined as the use of physiological or behavioral characteristics to recognize or verify the claimed identity of an individual. The demand for security access control systems based on personal identification is strongly increasing and growing in many important applications. Several different biometric characteristics exist and are in use in various applications. Each biometric has its strengths and weaknesses, and the choice depends on the application. Main biometric characteristics include DNA, face, hand vein, fingerprint, hand geometry, iris, palmprint, retina, signature, and voice. Multimodal biometric systems, which use more than one independent source of information to recognize individuals in order to improve recognition accuracy, have been successfully developed as well and are more and more exploited in applications [1]. The most commonly used technology in biometrics is the optical one, but other techniques (capacitive, thermal and more recently ultrasonic) have been experimented. Ultrasounds have some intrinsic advantages with respect to other methods; in fact they are not sensitive to the surface contaminations, such as stain, dirt, oil, ink and, more important, they provide information not only of the skin surface but also of the volume under the investigated skin region. Furthermore, ultrasounds detect life (Doppler mode) and therefore they can immediately detect fakes. Ultrasounds have already been exploited in biometrics applications. The main explored ultrasonic technique for livescan fingerprint consists in a XY mechanical scan [2-3]: a high frequency (high resolution) A-Scan ultrasonic transducer scans each dot of the XY area and stores in the memory all the reflected acoustic signals from each dot. The result is a 3D image of the under skin volume. The main drawback in the this technique relies in the very high scanning time, which could be prohibitive for practical applications. The scanning acquisition time can be reduced using an ultrasonic linear array transducer that moves in a direction perpendicular to the electronic scanning plane. This technique usually called 3D ultrasound [4-5], is a medical imaging modality with many potential applications that could be extended to biometrics as well. The authors has recently experimented this technique for the biometric recognition of hand geometry [6-7], palmprint [8-9] and fingerprint [9-11]. The data are acquired by the probe and stored in a series of B-scans that forms a 3D matrix representing a specific volume of the human hand. In this work a similar technique, which also exploit power Doppler analysis is employed to achieve 3D hand vein patterns. II. 2D CLASSICAL METHOD Palm vein recognition appeared in 1990s [12], and has been popular since 2000 because of the advantages it offers. A vein pattern reveals the vast network of blood vessels underneath a person’s skin. Like fingerprints, though it has never been proven in a strict scientific sense, the shape of the vascular patterns of different individuals are distinctive, and very stable over a long period of time. In addition, as the blood vessels are hidden underneath the skin and are mostly invisible to the human eye, vein patterns are much harder for intruders to copy compared to other biometric features. The properties of uniqueness, stability and strong immunity to forgery of the vein pattern make it a potentially good biometric which offers secure and reliable features for personal identity verification. The typical procedure used in palm vein biometrics is as follow (see Figure 1) [13]. The palm vein images are captured by a NIR camera as the input data. Then, a median filter is employed on the palm vein images to remove some noises and a suitable threshold is selected to segment the palm region. Finally, two baseline data points are found automatically and a rectangular region of interest (ROI) is obtained. 2-D Gabor filter are then normally used on the palm vein pattern to extract features. The Hamming distance is employed to 10.1109/ULTSYM.2012.0611