17-01-2013, 04:40 PM
Data Acquisition and Quality Analysis of 3-Dimensional Fingerprints
Data Acquisition and Quality Analysis.pdf (Size: 745.44 KB / Downloads: 25)
Abstract
This paper introduces a new technology of non-contact 3D fingerprint capture and processing for higher quality fingerprint
data acquisition. The system relies on a real-time 3D sensor using structured light illumination (SLI), which generates
both texture and detailed ridge depth information. The high resolution 3D scans are then converted into 2D flat equivalent
fingerprints. As a result, many limitations imposed upon conventional fingerprint capture and processing can be reduced by the
unobtrusiveness of this approach and the extra depth information acquired. The image quality of 2D flat equivalent fingerprint
is evaluated and analyzed using NIST fingerprint image software. A comparison is performed between the unraveled 3D
fingerprints and their 2D plain counterparts in terms of fingerprint quality.
INTRODUCTION
Fingerprint recognition has been extensively applied in both forensic law enforcement and security applications involving
personal identification [1]–[3]. Traditional fingerprint acquisition is performed in 2D using contact methods which have
evolved over the last century, from ink (rolled or plain) to capacitive, ultrasonic, pyroelectric, thermal, and optoelectronic
approaches [4]–[7]. Among fingerprint capture approaches, contact based devices detect the geometric difference between
contact and non-contact parts (i.e. ridges and valleys) of the finger on the device. The optical approach, on the other hand,
captures the texture information of the fingerprint under examination.
A typical automatic fingerprint identification system (AFIS) consists of four modules: image acquisition, preprocessing,
feature extraction, and feature matching [8], [9]. In the image acquisition, a digital image of the fingerprint is captured
from either an inked fingerprint impression, or electronic signals of a fingerprint sensor, or a live scan of the finger. The
preprocessing module assesses the quality of an acquired fingerprint and further enhances the acquired image for feature
extraction. Specific features are then extracted to represent the image in a certain feature space to facilitate matching. The
matching module computes the likelihood of the extracted feature set matching with a template feature set. The performance
of the system depends on how well each module performs and higher matching performance will be achieved if fingerprint
quality is sufficiently high.
In many applications that require high precision fingerprints, limitations are imposed upon the current fingerprint capture
technologies [10], [11], including:
1) obligatory maintenance of a clean sensor or prism surface;
2) uncontrollability and non-uniformity of the finger pressure on the device;
3) permanent or semi-permanent change of the finger ridge structure due to injuries or heavy manual labors;
4) residues from the previous fingerprint capture;
5) data distortion under different illumination, environmental, and finger skin conditions; and
6) extra scanning time and motion artifacts incurred in technologies that require finger rolling.
The majority of these limitations arise due to the physical contact of the finger surface with the sensor plate, or the nonlinear
distortion introduced by the 3D-to-2D mapping during image acquisition
Besides the robust operation, an ideal AFIS
system also requires automatic fingerprint entry, high-speed data acquisition, real-time feedback, and low cost. To address
these issues, several novel technologies have been developed [13]–[16] that avoid direct contact between the sensor and the
skin.
In [15], [16], Parziale et al proposed multi-camera touchless fingerprint scanner which acquires different finger views that
are combined together to provide a 3D representation of the fingerprint. Due to the lack of contact between the elastic skin
of the finger and any rigid surface, the acquired images preserve the fingerprints “ground-truth” without skin deformation
during acquisition [16]. However, employing the shape-from-silhouette scanning technique, the ridge information is obtained
from the surface reflection variation (i.e. albedo) information. Thus, the fingerprint is sensitive to surface color, surface
reflectance, geometric factors and some other effects.
We have been developing a non-contact 3D scanning system that employs structured light illumination (SLI). Our ultimate
goal is to simultaneously acquire 3D scans of all the five fingers and the palm in high speed and fidelity using multiple,
commodity digital cameras and a DLP projector. Post processing of these scans is then performed to virtually extract the
finger and palm surfaces, and create 2D flat equivalent images.
This work is partially funded by Flashscan3D, LLC, Richardson, TX and National Institute of Hometown Security, Somerset, KY.
The authors are with the Center for Visualization and Virtual Environments, University of Kentucky, Lexington, KY 40507. Qi Hao is now with
the Electrical and Computer Engineering Department of The University of Alabama. Laurence G. Hassebrook is a member of Flashscan3D, LLC.
Prototype scanner for 3D fingerprint scanning.
The advantages of the presented 3D fingerprint scanning and processing technology include:
1) non-contact defuses distortion that exists in conventional fingerprint acquisition system.
2) simultaneous acquisition of both texture and ridge depth information of fingers;
3) automated fingerprint entry in no need of interaction with the operator;
4) fast scanning (less than 1 second);
5) robustness to contamination of fingers and residues of previous users;
6) robustness to clutter and fraud because of the difficulties in faking 3D fingerprints;
7) real-time feedback (less than 5 seconds) for users to make position adjustment; and
8) low cost by using the off-the-shelf commodity camera and projector.
The rest of this paper is organized as follows. Section II describes the 3D acquisition setup. Section III presents the methods
to convert 3D scans to 2D flat equivalent finger images. Section IV describes the fingerprint quality evaluation methods.
Section V presents the experimental results. Section VI outlines future works.