Fingerprint identification method and apparatus

• US 4,185,270 A
• Filed: 06/14/1978
• Issued: 01/22/1980
• Est. Priority Date: 07/19/1976
• Status: Expired due to Term

• Current Assignee: Secure Technologies Group, Inc.
• Original Assignee: Secure Technologies Group, Inc.
• Inventors: Fischer, Frederic P. II, Pflueger, Phillip G.
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US05/915,475
• Time in Patent Office: 587 Days
• Field of Search: 340/146.3 E, 340/149 R
• US Class Current: 382/125
• CPC Class Codes: G06V 10/94 Hardware or software architectures specially adapted for image or video understanding
• CPC Class Codes: G06V 40/12 Fingerprints or palmprints

Abstract

A process and apparatus for encoding the ridge endings and bifurcations of an inquiry fingerprint and for automatically searching through a file of similarly encoded fingerprints for the fingerprint(s) whose minutiae are most similar to those of the inquiry fingerprint. The process does not require registration of the minutiae prior to storage or searching and the storage and search of file fingerprint cards can be implemented on a digital computer system since the minutiae data from each print is in a form easily processed by computer.

Claims

1. A method for automatically searching a computer readable file of minutiae-encoded fingerprints under the overall control of an electronic computer to identify for further processing those file fingerprints having the greatest degree of match with a search fingerprint comprising the steps of:

(a) entering data representing the locations and direction of the minutiae of the search fingerprint, said data being independent of any registration between the search and file fingerprints;

(b) matching the locations and direction of each and every one of the encodable minutiae of the search fingerprint contained in the entered data and the locations and direction of each and every one of the encoded minutiae of a file fingerprint;

(c) generating a set of registration parameters for each pair of search and file fingerprint minutiae matched;

(d) identifying any resulting clusters of said registration parameters;

(e) Developing a matching score based on the number of said clusters and the number of said registration parameters in said clusters;

(f) repeating steps (b) through (e) for at least a portion of the file of the minutiae-encoded fingerprints; and

(g) selecting those file fingerprints having the highest degree of match for further processing.

2. A method for automatically searching a computer readable file of minutiae-encoded fingerprints under the overall control of an electronic computer to identify for further processing those sets of file fingerprints having the greatest degree of match with a set of search fingerprints comprising the steps of:

(a) entering data representing the locations and direction of the minutiae of each of the fingerprints in the set of search fingerprints, said data being independent of any registration between the search and file fingerprints;

(b) matching the locations and direction of each and every one of the encodable minutiae of the set of search fingerprints against the locations and direction of each and every one of the encoded minutiae of corresponding ones of the fingerprints in a set of file fingerprints;

(c) generating a set of registration parameters for each pair of search and file fingerprint minutiae matched;

(d) identifying any resulting clusters of said registration parameters;

(e) developing a matching score based on the number of said clusters and the number of said registration parameters in said clusters;

(f) repeating steps (b) through (e) for at least a portion of the file of the set of minutiae-encoded fingerprints; and

(g) identifying those sets of file fingerprints in which at least one of the component fingerprints of a file set has more than a predetermined degree of match to a corresponding fingerprint of said set of search fingerprints.

3. The method of claim 2 wherein said step of identifying requires that more than one of the component fingerprints exceed said predetermined degree of match.

4. Apparatus for automatically searching a computer readable file of minutiae-encoded fingerprints under the overall control of an electronic computer to identify for further processing those file fingerprints having the greatest degree of match with a search fingerprint comprising:

encoding means for entering data representing the locations and direction of the minutiae of the search fingerprint, said data being independent of any registration between the search and file fingerprints;

comparison means for matching the locations and direction of each and every one of the encoded minutiae of the search fingerprint contained in the entered data and the locations and direction of each and every one of the encoded minutiae of a file fingerprint;

means to generate a set of registration parameters for each pair of search and file fingerprint minutiae matched;

means to identify any resulting clusters of said registration parameters; and

scoring means to provide a matching score based on the number of said clusters and the number of said registration parameters in said clusters.

5. Apparatus for automatically searching a computer readable file of minutiae-encoded fingerprints under the overall control of an electronic computer to identify for further processing those sets of file fingerprints having the greatest degree of match with a set of search fingerprints comprising the steps of:

encoding means for entering data representing the locations and direction of the minutiae of each of the fingerprints in the set of search fingerprints, said data being independent of any registration between the search and file fingerprints,

comparison means for matching the locations and direction of each and every one of the encodable minutiae of corresponding ones of the fingerprints in a set of file fingerprints;

means to generate a set of registration parameters for each pair of search and file fingerprint minutiae matched;

means to identify any resulting clusters of said registration parameters; and

scoring means to provide a matching score based on the number of said clusters and the number of said registration parameters in said clusters.