Image analysis – Applications – Personnel identification
Reexamination Certificate
1999-07-29
2003-04-08
Mehta, Bhavesh (Department: 2625)
Image analysis
Applications
Personnel identification
C382S125000, C382S209000
Reexamination Certificate
active
06546122
ABSTRACT:
CROSS REFERENCE TO APPENDIX
Appendix A, which is a part of the present disclosure, is a listing of software code for embodiments of components of this invention, which are described more completely below.
A portion of the disclosure of this patent document contains material which is subject to copyright protection. The copyright owner has no objection to the facsimile reproduction by anyone of the patent document or the patent disclosure, as it appears in the Patent and Trademark Office patent files or records, but otherwise reserves all copyright rights whatsoever.
FIELD OF THE INVENTION
This invention relates to fingerprint verification. More particularly, this invention relates to methods for combining fingerprint templates obtained from various portions of a finger into one fingerprint template representing the combined fingerprint.
BACKGROUND OF THE INVENTION
Biometric identification is used to verify the identity of a person by digitally measuring selected features of some physical characteristic and comparing those measurements with those filed for the person in a reference database, or sometimes on a smart card carried by the person. Physical characteristics that are being used include fingerprints, voiceprints, the geometry of the hand, the pattern of blood vessels on the wrist and on the retina of the eye, the topography of the iris of the eye, facial patterns and the dynamics of writing a signature and typing on a keyboard.
The fingerprint is one of the most widely used physical characteristics in biometric identification. Fingerprints are utilized as the most reliable means to identify individuals because of two outstanding characteristics, namely that they remain unchanged all through life and differ from individual to individual. Fingerprints consist of raised friction ridges of skin separated by recessed valleys of skin. Fingerprint “minutiae” are conventionally defined as ridge endings or ridge bifurcations where a ridge splits into two ridges.
Since the matching of a fingerprint as an image can require large file capacity to store the measured and referenced features and complex computation to match measured features with reference features, fingerprint identification is carried out using the positional relationship of features or minutiae that are extracted from the fingerprint image. The minutiae representing a fingerprint image are digitized and stored in a digital memory, which may be a read-only memory, magnetic tape, or the like. A fingerprint digitized in this manner may be compared with reference fingerprints stored in the memory. For the comparison to work, the reference fingerprint and the measured fingerprint must be extracted, characterized, and digitized so that the fingerprint templates contain the same information and have the same format.
As fingerprint sensing gains momentum in application, manufacturers are gearing up for high volume production, deploying both silicon-based sensors as well as the traditional optical sensors. As the production volumes increase, these manufacturers will continue to cost-reduce their products to stay competitive. One way to reduce cost is to decrease the size of the sensor. However, as the sensors get smaller, the matching accuracy is correspondingly reduced because less data is available for comparison. Furthermore, the size of the fingerprint sensor is often limited by the size of the finger in order to obtain a full fingerprint.
Placing the user's finger on the sensor in different positions and then pasting the images of the sensed portions of the finger together is one solution in reducing the size of a fingerprint sensor. The images taken from the various areas of the finger are typically aligned with a registration software and essentially “glued” together.
FIG. 1A
shows an image
10
having a partial fingerprint of a finger.
FIG. 1B
shows an image
12
having a partial fingerprint from the same finger but at a different position. As can be seen from
FIGS. 1A and 1B
, image
10
overlaps image
12
. For example, image
10
shows the top portion of a fingerprint while image
12
shows the bottom portion of the same fingerprint and both images
10
and
12
show the middle portion of the fingerprint.
FIG. 1C
shows an image
14
having a combined fingerprint derived by gluing images
10
and
12
together using an image correlation algorithm, for example, as described by Dana H. Ballard and Christopher M. Brown in a book entitled “Computer Vision” (1982) on pages 65-71, hereby incorporated by reference.
In image correlation, a template is created from a particular feature which is in effect a subimage. A similarity measure is computed which reflects how well the image data of another image match the template for each possible template location. The point of maximal match is then selected as the location of the feature. The template-matching calculations may be visualized by imagining the template being shifted across the image to different offsets to locate where in the image the feature occurs. The two images can then be aligned and combined.
However, image alignment, especially for poorly imaged fingers resulting from, for example, dry fingers, is a difficult task and is computationally expensive. Furthermore, because the finger has a fluid surface, the surface of the finger stretches and varies under different conditions. Hence, the ridge definition is rarely exactly the same twice. This inherent difference causes the overlap regions to be different, producing an artifact effect when glued together. In other words, the overlap region appears to be formed by two different sources because the transition at the overlap regions is not smooth. Therefore, the image correlation algorithm must take these conditions into account during the match, thereby requiring additional computation. A further problem is rotation which must be taken into account because the various portions of fingerprint may be obtained with the finger in different orientations.
Therefore, what is needed is a method for combining various templates from various portions of a finger, the method being computationally inexpensive.
SUMMARY OF THE INVENTION
A method for combining two templates. A first template having a first minutia set and a second template having a second minutia set are provided. In one embodiment, each template comprises at least one data chunk which represents one minutia. Each minutia is characterized by a location, a minutia angle and a neighborhood. The location further includes an x-coordinate and a y-coordinate. The neighborhood includes a plurality of neighbors, each having a distance from the selected minutia and two related angles. In one embodiment, all the parameters are quantized to reduce the bits required to represent the distance and the related angles of a minutia, hence decreasing the memory requirement.
The first template is compared with the second template to obtain a matching minutia set. The templates are compared on a data chunk by data chunk basis and by comparing each characterization parameters of the data chunks. Comparing on a data chunk by data chunk basis allows a computer with low computational power to execute the template matching algorithm. If all characterization parameters match, the data chunks match. The matching minutia pair is stored as a part of a matching minutia set in a memory accessible by a computer.
Registration parameters are calculated from the matching minutia set. Each registration parameter is calculated by summing the difference of each minutia pair in the matching minutia set and dividing the sum by the number of minutia pairs in the matching minutia set. In one embodiment, the registration parameters are in reference to the first minutia set. In another embodiment, the registration parameters are in reference to the second minutia set.
All minutiae in the minutia set that is not the reference minutia set are translated to be on the same coordinate system as the reference minutia set. The translation is done by adding the registration parameters to the characteri
Chawan Sheela
Mehta Bhavesh
Veridicom, Inc.
LandOfFree
Method for combining fingerprint templates representing... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method for combining fingerprint templates representing..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method for combining fingerprint templates representing... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3087763