A Multispectral Whole-Hand Biometric Authentication System

Abstract

We describe the design and development of a prototype whole-hand imaging system. The sensor is based on multispectral technology that is able to provide hand shape, fingerprints and palmprint modalities of a user's hand by a single user interaction with the sensor. A clear advantage of our system over other unimodal sensors for these modalities include: (i) faster acquisition time, (ii) better quality images, and (iii) ability to provide spoof detection. Initial results on a medium-size database show good recognition performance based on individual modalities as well as after fusing multiple fingers and fusing finger and palm. The prototype is being refined in order to improve performance even further.

Personal Authentication Using Finger Knuckle Surface

“…The palmprint and fingerprint features can be simultaneously extracted from the palm side hand images and combined to achieve the performance improvement [19], [30]. However, the size of finger knuckle is very small as compared with the palmprint and offers more attractive alternative as it also requires less processing as compared to palmprint.…”

Personal Recognition Using Single-Sensor Multimodal Hand Biometrics

“…However, their results only refer to randomly paired samples from multiple databases indicating an improvement of the best single biometric trait by means of Equal Error Rate (EER) performance of roughly 40% (yielding a total EER of 3.53%). Recently, an implemented solution extracting samples out of a single acquired input signal based on multispectral image processing has been presented by Rowe et al [3]. They report no errors on their dataset of 50 volunteers for fingerprint score fusion using the Sum Rule [4] and a significant performance improvement if ring finger and palmprint scores are combined.…”

“…While sensing regions may also be defined for other parts of the human hand (see [3] for results on metacarpal skin), we extract each fingerprint image as a w × h 3 (and w × h 2 for the thumb, respectively) rectangular area aligned with respect to the major axis of the finger circumscribed by its w × h sized bounding box. Since flatbed optical images exhibit low contrast between ridges and valleys, Contrast-limited adaptive histogram equalization described in [15] has been employed.…”

Contour Code: Robust and efficient multispectral palmprint encoding for human recognition

“…MSI has been extensively used in the fields of remote sensing, medical imaging and computer vision to analyze information in multiple bands of the electromagnetic spectrum. In the past decade, biometrics such as the face [2,3], iris [4] and fingerprints [5] have been investigated using multispectral images for improved human recognition. Recently, there has been an increasing interest in multispectral palmprint recognition.…”