Authentication method, and program and apparatus therefor

• US 20030161502A1
• Filed: 10/28/2002
• Published: 08/28/2003
• Est. Priority Date: 02/25/2002
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Fujii, Yusaku, Morihara, Takashi
• Granted Patent: US 7,079,671 B2
• Time in Patent Office: Days
• US Class Current: 382/115
• CPC Class Codes: G06V 40/1335 Combining adjacent partial images (e.g. slices) to create a composite input or reference pattern; Tracking a sweeping finger movement

Abstract

Section images of a human body part are continuously input by relative movement with respect to a read unit in an image input step, a living body section image including a character shape from an input section image group in a character area section step, and, when living body section images which include character shapes which are chronologically continuous are selected, the living body section images are connected to each other to generate a connected living body image in an image connection step. In the collation step, the connected living body image including the character shapes and living body section images which include character shapes which are not connected to each other are collated with a corresponding portion in reference information of a human body part which is registered in advance to authenticate an identical person.

Claims

1. An authentication method using living body information, comprising:

the image input step of continuously inputting section images of a human body part by relative movement with respect to a read unit;

the character area selection step of selecting a living body section image which includes a character shape from an input section image group;

the image connection step of, when living body section images which include the character shapes which are chronologically continuous are selected, connecting both the living body section images to each other to generate a connected living body image; and

the collation step of collating the connected living body image including the character shapes and living body section images which include character shapes which are not connected to each other with a corresponding portion in reference information of a human body part which is registered in advance to authenticate an identical person.

2. A method according to claim 1, wherein the character area selection step extracts, as the character shapes of the living body section images, any one of end points and branch points of fingerprint ridges, fingerprint ridge edge lines, portions where changes of tangent directions of the fingerprint ridge edge lines are large, or a combination thereof when the living body section images are fingerprint section images.

3. A method according to claim 1, wherein, in the character area selection step, the character shapes of the respective living body section images are detected, and the number of character shapes is counted, and a living body section image having character shapes the number of which is not less than a predetermined number is selected.

4. A method according to claim 1, wherein, in the image connection step, common areas are searched by correlative calculation of the character shapes in overlapping areas while relatively moving two living body section images which are chronologically continuous to detect relative amounts of movement of the images, and the two living body section images are connected to each other by using the detected amounts of movement.

5. A method according to claim 1, wherein the collation step comprises:

the moving amount detection step of detecting an amount of movement of a living body section image which does not include the character shape with respect to a living body section image which includes the character shape; and

the collation reference extraction step of extracting a portion used in collation from the living body reference information on the basis of the amount of movement.

6. A method according to claim 5, wherein, in the moving amount detection step, common areas are searched by correlative calculation of edge lines in overlapping areas while relatively moving two living body section images which are chronologically continuous to detect relative amounts of movement of the images, and a sum of amounts of movement is calculated depending on the number of continuous living body section images which do not include character shapes.

7. A method according to claim 1, wherein

the character area selection step selects some areas which include the character shapes from a living body section image which includes the character shape, and

the image connection step connects the some areas which include the character shapes and which are selected from living body section images which are chronologically continuous to generate a connected living body image.

8. A program for authenticating an identical person, causing a computer to execute:

the image input step of continuously inputting section images of a human body part by relative movement with respect to a read unit;

the character area selection step of selecting a living body section image which includes a character shape from an input section image group;

the image connection step of, when living body section images which include character shapes which are chronologically continuous are selected, connecting both the living body section images to each other to generate a connected living body image; and

the collation step of collating the connected living body image including the character shapes and living body section images including the character shapes which are not connected to each other with a corresponding portion in reference information of a human body part which is registered in advance to authenticate an identical person.

9. A program according to claim 8, wherein the character area selection step extracts, as the character shapes of the living body section images, any one of end points and branch points of fingerprint ridges, fingerprint ridge edge lines, portions where changes of tangent directions of the fingerprint ridge edge lines are large, or a combination thereof when the living body section images are fingerprint section images.

10. A program according to claim 8, wherein, in the character area selection step, the character shapes of the respective living body section images are detected, and the number of character shapes is counted, and a living body section image having character shapes the number of which is not less than a predetermined number is selected.

11. A program according to claim 8, wherein, in the image connection step, common areas are searched by correlative calculation of the character shapes in overlapping areas while relatively moving two living body section images which are chronologically continuous to detect relative amounts of movement of the images, and the two living body section images are connected to each other by using the detected amounts of movement.

12. A program according to claim 8, wherein the collation step comprises:

the moving amount detection step of detecting an amount of movement of a living body section image which does not include the character shape with respect to a living body section image which includes the character shape; and

the collation reference extraction step of extracting a portion used in collation from the living body reference information on the basis of the amount of movement.

13. A program according to claim 12, wherein, in the moving amount detection step, common areas are searched by correlative calculation of edge lines in overlapping areas while relatively moving two living body section images which are chronologically continuous to detect relative amounts of movement of the images, and a sum of amounts of movement is calculated depending on the number of continuous living body section images which do not include character shapes.

14. A program according to claim 8, wherein

the character area selection step selects some areas which include the character shapes from a living body section image which includes the character shape, and

the image connection step connects the some areas which include the character shapes and which are selected from living body section images which are chronologically continuous to generate a connected living body image.

15. An authentication apparatus using living body information, comprising:

an image input unit for continuously inputting section images of a human body part by relative movement with respect to a read unit;

a character area selection unit for selecting a living body section image which includes a character shape from an input section image group;

an image connection unit for, when living body section images which include character shapes which are chronologically continuous are selected, connecting both the living body section images to each other to generate a connected living body image; and

a collation unit for collating the connected living body image including the character shapes and living body section images including character shapes which are not connected to each other with a corresponding portion in reference information of a human body part which is registered in advance to authenticate an identical person.