Skin-pattern recognition method and device

• US 4,805,223 A
• Filed: 12/18/1986
• Issued: 02/14/1989
• Est. Priority Date: 04/22/1985
• Status: Expired due to Term

• Current Assignee: VLSI Vision Ltd.
• Original Assignee: The Quantum Fund Ltd.
• Inventors: Denyer, Peter B.
• Primary Examiner(s): Boudreau, Leo H.
• Application Number:
  US07/002,402
• Time in Patent Office: 789 Days
• Field of Search: 382/4, 382/5, 382/52, 356/71
• US Class Current: 382/127
• CPC Class Codes: G06V 40/12 Fingerprints or palmprints
• CPC Class Codes: G06V 40/1365 Matching; Classification
• CPC Class Codes: G07C 9/257 electronically G07C9/26 takes precedence
• CPC Class Codes: G07C 9/37 using biometric data, e.g. fingerprints, iris scans or voice recognition

Abstract

A method of authenticating a person by comparing information from a currently taken skin-pattern with stored information from a previously taken skin-pattern is characterized in that the currently taken skin-pattern is produced on a contact surface, an image of the print of the skin pattern on the contact surface is projected onto a photo-detector device (20), the intensity variations of the print image in at least one region thereof impinging on the photo-detector device is assessed in digital form in a multiplicity of different sub-regions of the or each said region to create (in unit 21) a first digital signal train, a second digital train is derived from the stored skin pattern information (in store 23) and the first and second signal trains are compared (in 24) to determine the degree of coincidence therebetween, an assessment of the determined degree being used to decide (on indicator 27) whether the currently-taken and stored skin patterns are sufficiently similar to authenticate the person.

Claims

1. A method of authenticating a person by comparing information from a currently taken skin pattern with stored information from previously taken skin pattern, the currently taken skin-pattern being produced on a contact surface, an image of the print of the skin pattern on the contact surface being projected into a photo-detector device, the intensity variations of the image of the skin pattern in at least one region thereof impinging on the photo-detector device being assessed in digital form in a series of binary "1" units and "0" units to create a first digital signal train by setting a threshold intensity and judging image areas of intensity above the threshold as meriting one type of binary unit and other image areas of intensity below the threshold as meriting the other type of binary unit, a second digital signal train being derived from the stored skin pattern information and the first and second signal trains being compared to determine the degree of coincidence therebetween, an assessment of the determined degree being used to decide whether the currently taken and stored skin patterns are sufficiently similar to authenticate the person, the degree of coincidence between the first and second signal trains being assessed by looking for correlation between digital signal trains representative of respective first sub-regions of each of the currently taken and stored skin patterns and between respective second sub-regions of each of the currently taken and stored skin patterns while allowing relative translational and rotational movements between the first and second sub-regions of one of the skin patterns, the threshold used for the first digital signal train being adapted according to an average value of intensity within a sampling region of the skin pattern image used to create the first digital signal train, and the second signal train being derived by considering only determinate samples of the previously taken skin pattern image which can clearly be categorized as a binary "1" or a binary "0" so that in the correlation between the first and second signal trains, all "indeterminate" samples in the second signal train are discounted.

2. A method as claimed in claim 1, characterized in that a correlation score is first computed using just first sub-regions of each skin pattern and the second sub-regions are used only if a predetermined correlation score is exceeded with the first sub-regions and in that authentication is determined only if a predetermined correlation score is exceeded with both first and second sub-regions.

4. In the operation of the method of claim 1, the use of a fingerprint capture device comprising a photo-detector device having a surface divided into a matrix of separate photosensitive areas, each of which can monitor the light intensity of one pixel of an image projected thereon and an optical system including a lens having an optical axis to project a fingerprint pattern onto said matrix-divided surface, the fingerprint pattern being created on the hypotenuse face of a right angle prism, characterized in that the prism, an illuminating source for the hypotenuse face, the optical system for projecting an image of the hypotenuse face onto said matrix-divided surface, and the photo-detector device are mounted together on a circuit board with the optical axis of the lens extending parallel to a surface of the board to define a compact composite unit.

5. A device as claimed in claim 4, characterized in that the hypotenuse face of the prism includes means to encourage consistently similar locations of fingers on the hypotenuse face.

6. A device as claimed in claim 4, character in that the device is used to create a first signal train from a real time image of a fingerprint of a presenter of a sheet member required for a cash transaction and is associated with means to read from the sheet member the second signal train.

7. A device as claimed in claim 1, characterized in that a local capture device is used to create a first signal train from a real time image of a fingerprint of a person requiring verification, the local capture device being inked to a remote data bank from which the second signal rains of a fingerprints of many different persons can be generated and the local capture device being associated with means whereby a nominated person'"'"'s second signal train can be called up for comparison with an output of the local capture device.

8. A device as claimed in claim 4, characterized in that it is associated with the release mechanism of a normally locked site access device.

3. A method as claimed in claim 14, characterized in that third and subsequent sub-regions of each skin pattern are used if a predetermined correlation score is exceeded with the first and second sub-regions and authentication is determined only if a predetermined correlation score is exceeded with all sub-regions used.