Falsification discrimination method for iris recognition system

• US 6,760,467 B1
• Filed: 03/21/2000
• Issued: 07/06/2004
• Est. Priority Date: 03/23/1999
• Status: Expired due to Term

• Current Assignee: Iris Id
• Original Assignee: LG Electronics, Inc. (LG Corporation)
• Inventors: Chae, Jang Jin, Min, Seung Gi
• Primary Examiner(s): Patel, Jayanti K.
• Application Number:
  US09/532,042
• Time in Patent Office: 1,568 Days
• Field of Search: 382/103, 382/117, 382/190, 382/115, 382/191, 382/155, 382/209, 382/291, 382/312, 382/317, 382/324, 351/221, 351/210, 351/209, 351/206, 351/205, 396/51, 396/50, 396/123, 396/57, 396/18, 434/155, 434/159, 434/271, 348/139, 348/335
• US Class Current: 382/117
• CPC Class Codes: G06V 40/18 Eye characteristics, e.g. of the iris
• CPC Class Codes: G06V 40/40 Spoof detection, e.g. liveness detection

Abstract

The present invention relates to a technique for checking if the reflected image of a light emitting diode for lighting is detected in the image of an iris and determining whether the iris is a living one in order to prevent an illegal intrusion using a simulacrum such as a photograph or dynamic image. The object of the present invention is achieved by a first step of imaging an eye of the user when the user approaches the field of view, processing the image signal, and capturing the image of the iris and pupil, a second step of selectively lighting LEDs for lighting installed on the both sides of a camera while imaging the eye of the user and detecting the reflected image of a corresponding LED in the image of the iris and pupil, a third step of determining that the iris is a falsification and informing the system of the fact if the reflect image of a the corresponding light emitting diode is not detected in the above steps, and determining that the iris is really recognized if detected, and a fourth step of performing a recognition of an iris image signal using a database if the iris is proven to be really recognized and determining whether the image is data registered in the database.

Claims

1. A falsification discrimination method for an iris recognition system, comprising:

a first step of imaging a user'"'"'s eye when the user approaches a field of view, processing the image signal, and determining whether the image is precisely focused;

a second step of selectively lighting a light emitting diode installed near a camera during imaging while checking if the reflected image of a corresponding light emitting diode is presented in the image of the iris and pupil; and

a third step of determining that the iris is a falsification and informing the system of the fact if the reflected image of the corresponding light emitting diode is not detected as the result of the comparison, and determining whether the iris is really recognized if the image is detected, wherein the selectively lighting of the light emitting diode comprises lighting at least two distinct light sources, and wherein the determining that the iris is a falsification comprises determining whether the reflected image comprises images of reflections of sequential lighting of the light emitting diodes corresponding to images of predetermined reflections of the sequential lighting of the light emitting diodes.

2. The method of claim 1, wherein, in the third step of determining whether the iris is not recognized, the confirmation of a living iris is informed if the iris is recognized using a database, and the rejection of a non-living iris is informed if the iris is not recognized.

3. The method of claim 1, wherein the second step is performed by sequentially lighting light emitting diodes for lighting installed near a camera.

4. The method of claim 1, wherein the second step is performed by lighting all the light emitting diodes for lighting installed near the camera.

5. The method of claim 1, wherein said selectively lighting a light emitting diode comprises lighting at least two distinct lighting sources during at least two different predetermined times that reflect off the user'"'"'s eye as the reflected image.

6. The method of claim 5, wherein the determining that the iris is a falsification comprises determining whether the reflected image of the at least two distinct lighting sources are reflected correctly off the user'"'"'s eye.

7. The method of claim 1, wherein the falsification discrimination method distinguishes actual, live irises of eyes from images of eyes including photographs of irises or dynamic images of irises.

8. A falsification discrimination method for an iris recognition system, comprising:

a first step of imaging a user'"'"'s eye when the user approaches a field of view, processing the image signal, and capturing the image of an iris and pupil;

a second step of selectively lighting a fight emitting diode installed near a camera during imaging while checking if the reflected image of a corresponding light emitting diode is presented in the image of the iris and pupil; and

a third step of determining whether the iris image is data registered in a database and informing the system of the rejection of a non-living iris if the iris is not really recognized and determining whether the iris is a falsification if the iris is really recognized, wherein the selectively lighting of the light emitting diode comprises lighting at least two distinct light sources, and wherein the determining that the iris is a falsification comprises determining whether the reflected image comprises images of reflections of sequential lighting of the light emitting diodes corresponding to images of predetermined reflections of the sequential lighting of the light emitting diodes.

9. The method of claim 8, wherein, in the third step of determining whether the iris is a falsification, a falsification is informed if the reflected image of the light emitting diode is not presented in the second step and the confirmation of a living iris is informed if the iris is not a falsification.

10. The method of claim 8, wherein the second step is performed by sequentially lighting light emitting diodes for lighting installed near a camera.

11. The method of claim 8, wherein the second step is performed by lighting all the light emitting diodes for lighting installed near the camera.

12. The method of claim 8, wherein selectively lighting a light emitting diode comprises lighting at least two distinct lighting sources during at least two different predetermined times that reflect off the user'"'"'s eye as the reflected image.

13. The method of claim 12, wherein the determining that the iris is a falsification comprises determining whether the reflected image of the at least two distinct lighting sources are reflected correctly off the user'"'"'s eye.

14. The method of claim 8, wherein the falsification discrimination method distinguishes actual, live irises of eyes from images of eyes including photographs of irises or dynamic images of irises.

15. In an iris recognition system, a falsification discrimination apparatus for an iris recognition system, comprising:

a camera for imaging the eye of a person attempting to be recognized;

an iris image processing unit for capturing the image of the iris and pupil by processing an output image signal of the camera, outputting a lighting control signal of light emitting diodes while confirming whether a LED image is presented on a corresponding image in the pupil image, and determining whether the iris image is true or false;

a light emitting diode drive unit for selectively lighting the light emitting diodes for lighting installed at the right and left sides of the camera from the eye according to the lighting control signal;

a distance measurement unit for measuring the distance from the camera to the eye utilizing infrared light; and

an iris recognition unit for performing an iris recognition process using a database when it is confirmed that the iris is imaged as the result of the comparison, and determining that the iris image is a falsification by determining whether the reflected image of a corresponding light emitting diode comprises images of reflections of sequential lighting of the light emitting diodes corresponding to images of predetermined reflections of the sequential lighting of the light emitting diodes.

16. The apparatus of claim 15, further comprising light emitting diodes that emit light from at least two distinct light sources, wherein the iris image processing unit confirms that a reflection of the at least two distinct light sources are correctly reflected on said iris in comparison to a predetermined reflection of the at least two distinct light sources.

17. The apparatus of claim 16, wherein the at least two distinct light sources emit light during at least two predetermined times and said reflection of the at least two distinct light sources are confirmed corresponding to said at least two predetermined times.

18. The apparatus of claim 16, wherein said reflection of the at least two distinct light sources are reflected off the surface of the eye of the person attempting to be recognized.

19. In an iris recognition system, a falsification discrimination apparatus for an iris recognition system, comprising:

a camera for imaging the eye of a person attempting to be recognized;

an iris image processing unit for capturing the image of the iris and pupil by processing an output image signal of the camera, outputting a lighting control signal of light emitting diodes while confirming whether a LED image is presented on a corresponding image in the pupil image, and determining whether the iris image is true or false;

a light emitting diode drive unit for selectively lighting the light emitting diodes for lighting installed at the right and left sides of the camera from the eye according to the lighting control signal;

a distance measurement unit for measuring the distance from the camera to the eye utilizing infrared light;

an iris recognition unit for performing an iris recognition process using a database when it is confirmed that the iris is imaged as the result of the comparison; and

an informing unit for informing the system of iris recognition and falsification, and determining that the iris image is a falsification by determining whether the reflected image of a corresponding light emitting diode comprises images of reflections of sequential lighting of the light emitting diodes corresponding to images of predetermined reflections of the sequential lighting of the light emitting diodes.

20. The apparatus of claim 19, further comprising light emitting diodes that emit light from at least two distinct light sources, wherein the iris image processing unit confirms that a reflection of the at least two distinct light sources are correctly reflected on said iris in comparison to a predetermined reflection of the at least two distinct light sources.

21. The apparatus of claim 20, wherein the at least two distinct light sources emit light during at least two predetermined times and said reflection of the at least two distinct light sources are confirmed corresponding to said at least two predetermined times.

22. The apparatus of claim 20, wherein said reflection of the at least two distinct light sources are reflected off the surface of the eye of the person attempting to be recognized.

23. A method, comprising:

lighting an eye using a first distinct light source;

processing a first iris image of said eye including a first reflection of said first distinct light source off said eye;

detecting a first pupil position of said eye and a first reflection position of said first distinct light source on said eye;

determining whether said first reflection position is correct;

lighting said eye using a second distinct light source in a position different from said first distinct light source;

processing a second iris image of said eye including a second reflection of said second distinct light source off said eye;

detecting a second pupil position of said eye and a second reflection position of said second distinct light source on said eye; and

determining whether said second reflection is correct, wherein if said first and said second reflections are correct, then an iris recognition is performed.

24. The method of claim 23, wherein determining whether said first and said second reflection positions are correct comprises comparing said reflection positions corresponding to said first and second distinct light sources to predetermined positions of said first and second distinct light sources, wherein the method further comprises:

identifying a person after said first and second reflection positions are determined to be correct by identifying said eye in a database.

25. A falsification discrimination method for an iris recognition system, comprising:

selectively lighting light sources installed at the iris recognition system; and

determining that an iris image is a falsification by determining whether the reflected image of a corresponding light source comprises iris images of reflections of sequential lighting of the light sources corresponding to images of predetermined reflections of the sequential lighting of the light sources.

26. The method of claim 25, wherein a position of the corresponding light source is changed.