METHOD AND APPARATUS FOR DETECTING FORGED FACE USING INFRARED IMAGE

• US 20100128938A1
• Filed: 11/24/2009
• Published: 05/27/2010
• Est. Priority Date: 11/25/2008
• Status: Active Grant

• Current Assignee: Electronics and Telecommunications Research Institute
• Original Assignee: Electronics and Telecommunications Research Institute
• Inventors: Jung, Sung Uk, Lee, Yongjin, Chung, Yun Su, Moon, Ki Young
• Granted Patent: US 8,582,833 B2
• Time in Patent Office: Days
• US Class Current: 382/117
• CPC Class Codes: G06V 10/143 Sensing or illuminating at different wavelengths
• CPC Class Codes: G06V 40/166 using acquisition arrangements

Abstract

A method for detecting a forged face using an infrared image includes: acquiring an infrared image and a photorealistic image captured by one or two cameras; extracting a face region from the photorealistic image; determining, based on analysis of quality characteristics of the infrared image, whether a current face is a forged face or not; and performing face recognition on the extracted face region if it is determined that the current face is a non-forged face. The method further includes capturing an infrared image and a photorealistic image again through the camera without performing face recognition if it is determined that the current face is a forged face.

Claims

1. A method for detecting a forged face using an infrared image, comprising:

acquiring an infrared image and a photorealistic image captured by one camera or two cameras

extracting a face region from the photorealistic image;

determining, based on analysis of quality characteristics of the infrared image, whether a current face is a forged face or not; and

performing face recognition on the extracted face region if it is determined that the current face is a non-forged face.

2. The method of claim 1, further comprising:

capturing an infrared image and a photorealistic image again through the cameras without performing face recognition if it is determined that the current face is a forged face.

3. The method of claim 1, wherein said determining whether the current face is a forged face comprises:

evaluating the infrared image;

detecting whether a facial shape is present or not in the extracted face region by using pattern recognition technology; and

analyzing eyeball characteristics of the infrared image.

4. The method of claim 3, wherein the evaluating the infrared image comprises:

evaluating clearness of a region in the infrared image corresponding to the extracted face region by using sharpness technique;

determining the current face to be a forged face when, as a result of the evaluation, a clearness degree of the region in the infrared image is less than a preset clearness value; and

determining the current face to be a non-forged face when, as a result of the evaluation, the clearness degree of the region in the infrared image is identical to or greater than the preset clearness value.

5. The method of claim 3, wherein the determining whether the facial shape is present or not comprises:

detecting, based on characteristics of the infrared image indicating that a facial shape is not visualized in a case of the infrared image captured from the forged face, whether a facial shape is present or not by applying the pattern recognition technique to the face region;

determining the current face to be a forged face when, as a result of the detection, a facial shape is not present; and

determining the current face to be a non-forged face when, as a result of the detection, a facial shape is present.

6. The method of claim 5, wherein the pattern recognition technique uses Adaboost or a support vector machine.

7. The method of claim 3, wherein the analyzing the eyeball characteristics comprises:

comparing an eyeball region of the infrared image with preset eyeball regions;

determining the current face to be a forged face when, as a result of the analysis, the eyeball region of the infrared image does not have characteristics identical to those of the preset eyeball regions; and

determining the current face to be a non-forged face when, as a result of the analysis, the eyeball region of the infrared image has characteristics identical to those of at least one of the preset eyeball regions.

8. An apparatus for detecting a forged face using an infrared image, comprising:

an image acquisition unit for acquiring an infrared image and a photorealistic image captured by one or two cameras;

a face region extraction unit for extracting a face region from the photorealistic image;

a forged face determination unit for determining, based on analysis of quality characteristics of the infrared image, whether a current face is a forged face; and

a face recognition unit for performing face recognition on the extracted face region if it is determined that the current face is a non-forged face.

9. The apparatus of claim 8, wherein the forged face determination unit comprises:

a face image evaluation unit for evaluating clearness of a region in the infrared image corresponding to the extracted face region by using sharpness technique to determine the current face to be a forged face when a clearness degree of the region in the infrared image is less than a preset clearness value or to determine the current face to be a non-forged face when the clearness degree of the region in the infrared image is identical to or greater than the preset clearness value; and

a face presence detection unit for detecting, based on characteristics of the infrared image indicating that a facial shape is not visualized in a case of the infrared image captured from the forged face, whether a facial shape is present or not by applying the pattern recognition technique to determine the current face to be a forged face when a facial shape is not present or to determine the current face to be a non-forged face when a facial shape is present; and

an eyeball characteristic analysis unit for comparing an eyeball region of the infrared image with preset eyeball regions to determine the current face to be a forged face when the eyeball region of the infrared image does not have characteristics identical to those of the preset eyeball regions or to determine the current face to be a non-forged face when the eyeball region of the infrared image has characteristics identical to those of at least one of the preset eyeball regions.

10. The apparatus of claim 8, wherein when the forged face determination unit determines that the current face is a forged face an infrared image and a photorealistic image of the current face are captured again through the cameras without performing face recognition by face recognition unit.

11. The apparatus of claim 9, wherein the pattern recognition technique uses Adaboost or a support vector machine.