KR20150069799A - Method for certifying face and apparatus thereof - Google Patents

Abstract

The present invention relates to a method and a device for authenticating a face. According to the present invention, the method for authenticating a face includes: a step of separately obtaining a general image and a thermal image by separately photographing each random object by using a general camera and an infrared camera; a step of extracting a face area from the obtained thermal image through an image processing part; a step of determining whether there is thermal distribution in the extracted face area through a fake face determining part; and a step of detecting a fake face if there is no thermal distribution in the extracted face area after the determination, and detecting an actual face if there is the thermal distribution in the extracted face area. According to the present invention, as user authentication is performed by using the general image and the thermal image using the general camera and the infrared camera, difficulty in finding a face area from the existing camera image (visible light image) is solved, and the security and reliability of face authentication is improved.

Description

[0001] The present invention relates to a face authentication method and apparatus,

The present invention relates to a face authentication method and apparatus, and more particularly, to a face authentication method capable of detecting whether or not a false face is detected by analyzing characteristics of an infrared (thermal) image from a face image captured using an infrared ray and a visible ray camera And a device therefor.

Recognition technology using biometrics is collectively referred to as identification of individual's identity according to the physical or behavioral characteristics inherent in human beings. With the development of communication technology, various services through network have become common, and biometrics technology is one of the fields where research is concentrated on user authentication solution.

Such biometric recognition technology is used for face authentication, fingerprint authentication, iris authentication, and voice authentication as personal authentication techniques using biometric information such as personal body characteristics or behavior. Such bio information is characterized by the fact that there is no fear of losing and there is no need to memorize. In addition, because biometric technology requires individual biometric information to be input directly to the system, it has the advantage of being safer than the technology using existing passwords in terms of security. Accordingly, the existing biometric authentication technology based on the existing password and ID card- As a next-generation technology.

One of the problems faced in the conventional face authentication system is to print a photograph of a person other than an actual face in a real size and recognize it in the light of the camera. Therefore, the system needs an additional device to determine whether it is a real face or a photograph.

In order to cope with this, conventionally, a method of measuring the three-dimensional change appearing on the face by illuminating the side from the side is used. However, it was difficult to put into practical use due to the user's eyes being scratched or an installation problem in the interior. In addition, a method of discriminating the subject by blinking the eyes has been proposed, but this also causes the user to feel discomfort and inconvenience.

In addition, the biometric authentication using the face so far has been carried out using an image (image) photographed by visible light, so authentication accuracy is lowered. In addition, when performing face authentication using a visible light image, since the image processing must be performed to search for the face area for the whole image area in order to find the face area in the image captured by the camera, it takes a lot of calculation time and time. In addition, when the illuminance difference is large due to the influence of ambient light and background, the face authentication method using visible light has a problem of low accuracy in finding a face area.

Japanese Unexamined Patent Application Publication No. 2008-059408 United States patent publication US 2010/0134250

The present invention has been made in view of the above-mentioned problems, and it is an object of the present invention to provide a method and apparatus for simultaneously capturing facial images using infrared and visible light cameras having different angles of view and analyzing characteristics of infrared (thermal) The present invention has been made in view of the above problems.

According to an aspect of the present invention,

a) capturing an arbitrary object by using a general camera and an infrared camera, respectively, to obtain a general image and a thermal image;

b) extracting a face region from the obtained thermal image by an image processing unit;

c) determining whether or not a thermal distribution exists in the extracted face area by the fake face determination unit; And

d) if the extracted face region does not have a heat distribution, it is detected as a false face, and if there is a thermal distribution in the extracted face region, the detected face region is detected as an actual face .

The method may further include converting the extracted face region into a face region based on the general image based on the extracted face region using the information on the extracted face region in the step b) can do.

If the face is detected as a forged face in the step d), a warning light or a warning sound indicating a forged face may be generated by the forged face determination unit or the face authentication unit.

The step of calculating the size of the facial feature part with respect to the extracted facial area using the facial area conversion value for the general image by the facial feature size calculating part when the facial feature size is detected in step d) .

The method may further include transmitting facial feature size data calculated by the facial feature size calculating unit to the face authenticating unit and comparing the facial feature size data with preset reference data to authenticate the face as an actual face.

According to another aspect of the present invention, there is provided a face authentication apparatus comprising:

A photographing unit for photographing a subject in a visible light wavelength region and an infrared light wavelength region, respectively;

An image acquiring unit for acquiring a general image and a thermal image photographed by the photographing unit, respectively;

An image processing unit receiving the general image and the thermal image acquired by the image acquiring unit and extracting the face area, the face area conversion, and the size of the facial feature part;

A fake face determination unit for determining whether a face is forged based on a result of face region extraction from the image processing unit; And

And a face authentication unit for receiving the processing results from the image processing unit and the counterfeit face determination unit and authenticating the face, respectively.

Here, the photographing unit may include a general camera for photographing a subject in a wavelength region of visible light, and an infrared camera for photographing in an infrared wavelength region.

The image acquiring unit may include a general image acquiring unit for acquiring a general image photographed by a general camera of the photographing unit in a wavelength region of visible light, a column for acquiring a thermal image photographed in an infrared wavelength region by the infrared camera of the photographing unit, And an image acquisition unit.

The image processing unit may further include a face region extracting unit for extracting a face region from the thermal image received from the thermal image obtaining unit of the image obtaining unit, A face region conversion unit for converting the face region extracted by the extraction unit into a face region based on a general image, and a face feature extraction unit for calculating a face feature size from a face region based on a general image converted by the face region conversion unit, And a size calculating unit.

In addition, the forgery face determination unit can determine whether or not a forged face exists by determining whether or not a thermal distribution exists in the face region extracted by the face region extraction unit.

According to the present invention, user authentication is performed using a thermal image and a general image using an infrared camera and a general camera, thereby solving the difficulty in finding a face area in a conventional general camera image (visible light image) And it is possible to improve reliability and security in face authentication.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a view schematically showing a configuration of a face authentication apparatus according to an embodiment of the present invention; FIG.
FIG. 2 is a flowchart illustrating an execution process of a face authentication method according to an embodiment of the present invention; FIG.
3 is a diagram schematically illustrating a concept of shooting by a general camera and an infrared camera different in angle of view, which is employed in the face authentication device shown in Fig.
4 is a view showing a concept of photography by a general camera and an infrared camera involved in face area conversion in the face authentication method according to the present invention.
5 is a diagram schematically illustrating the concept of converting a face region extracted from an infrared camera image into a face region based on a general camera image in the face authentication method according to the present invention.
FIG. 6 is a view showing a face area based on a converted general camera image in a resolution-converted image in the face authentication method according to the present invention.
7 is a diagram schematically illustrating the concept of calculating the size of a facial feature in the face authentication method according to the present invention.

The terms and words used in the present specification and claims should not be construed as limited to ordinary or dictionary terms and the inventor can properly define the concept of the term to describe its invention in the best way Should be construed in accordance with the principles and meanings and concepts consistent with the technical idea of the present invention.

Throughout the specification, when an element is referred to as "comprising ", it means that it can include other elements as well, without excluding other elements unless specifically stated otherwise. Also, the terms " part, "" module, "and" device " Lt; / RTI >

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a view schematically showing a configuration of a face authentication apparatus according to an embodiment of the present invention.

1, a face authentication apparatus 100 according to the present invention includes a photographing unit 110, an image obtaining unit 120, an image processing unit 130, a counterfeit face determination unit 140, a face authentication unit 150, As shown in FIG.

As shown in Fig. 3, the photographing unit 110 photographs a subject in a visible light wavelength region and an infrared light wavelength region, respectively. Here, the photographing unit 110 may include a general camera 111 and an infrared camera 112. The general camera 111 photographs a subject in a visible light wavelength region, and provides the photographed image to a general image acquiring unit 121, which will be described later. The infrared camera 112 photographs the subject in the infrared wavelength range, and provides the photographed image to the thermal image acquiring unit 122, which will be described later. As shown in FIG. 3, the above-described photographing unit 110 can photograph a wider area when the infrared camera is wider than the general camera (A > B). When the infrared camera is wider than the general camera, the infrared camera can be configured to operate only the infrared camera and the general camera to wait in the power saving mode in driving the camera. Thus, when one of the two cameras is configured to wait in the power saving mode, there is an advantage that unnecessary power consumption can be reduced.

3, when the face of the person is detected in the primary sensing area, the general camera is operated. When the face is detected in the recognition area, the face authentication unit 150 May be configured to perform a subsequent operation such that the face authentication is performed.

The image acquiring unit 120 acquires a general image and a thermal image photographed by the photographing unit 110 described above. The image acquisition unit 120 may include a general image acquisition unit 121 and a thermal image acquisition unit 122. The general image acquiring unit 121 acquires a general image photographed in a visible light wavelength region by the general camera 111 of the photographing unit 110 and acquires a general image in a form suitable for image processing in an image processing unit 130 And provides the modified image to the image processing unit 130. The thermal image obtaining unit 122 obtains the thermal image photographed in the infrared wavelength region by the infrared camera 112 of the photographing unit 110 and obtains the thermal image in a form suitable for the image processing in the image processing unit 130, And provides the modified image to the image processing unit 130.

The image processor 130 receives the general image and the thermal image obtained by the image acquiring unit 120, and calculates face area extraction, face area conversion, and size of the facial feature. The image processing unit 130 may include a face area extracting unit 131, a face area converting unit 132, and a facial feature size calculating unit 133.

The face region extracting unit 131 extracts a face region from the thermal image received from the thermal image obtaining unit 122 of the image obtaining unit 120. The face region transforming unit 132 transforms the face region extracted by the face region extracting unit 131 based on the general image received from the general image obtaining unit 121 of the image obtaining unit 120, Of the face area. The facial feature size calculating unit 133 calculates facial feature size from a face image area based on the general image converted by the face area converting unit 132. [

The fake face determination unit 140 determines whether or not the face is forged based on the face region extraction result from the face region extraction unit 131 of the image processing unit 130. [ Such a forged face determiner 140 can determine whether or not a forged face exists by determining whether or not there is a thermal distribution in the face area extracted by the face area extractor 131, for example. That is, if there is a thermal distribution in the extracted face region, it is determined that the face is not a forged face, that is, it is an actual face, and if there is no heat distribution in the extracted face region, it is discriminated as a forged face.

The face authentication unit 150 receives the processing results from the image processing unit 130 and the fake face determination unit 140 to perform face authentication. That is, when the fake face determination unit 140 receives the fake face determination result, the face authentication unit 150 may authenticate that the face is a forged face and display a message therefor. In addition, a warning light may be displayed or a warning sound may be generated. The facial authentication unit 150 receives facial feature size data calculated by the facial feature size calculator 133 of the image processor 130 and stores facial feature size data, If the comparison result is within the allowable error range, it is authenticated as an actual face.

Hereinafter, a method for authenticating a face using the face authentication apparatus according to the present invention having the above-described configuration will be described.

2 is a flowchart illustrating a process of executing a face authentication method according to an embodiment of the present invention.

Referring to FIG. 2, according to the face authentication method according to the present invention, first of all, an arbitrary object is photographed using a general camera 111 and an infrared camera 112, respectively, to acquire a general image and a thermal image S201 to S203). That is, by using the general camera 111 and the infrared camera 112 of the photographing unit 110, an arbitrary object is photographed in the visible light wavelength region and the infrared light wavelength region, respectively, 121 and the thermal image acquiring unit 122, respectively, thereby acquiring a general image and a thermal image, respectively.

When the general image and the thermal image captured by the general camera 111 and the infrared camera 112 are acquired as described above, the face region extracting unit 131 of the image processing unit 130 extracts the face region (Step S204).

At this time, preferably, the face area conversion unit 132 converts the extracted face area into a general image-based face area based on the general image, using the information about the face area extracted in step S204 Step S205 may be further included.

Here, the facial area conversion unit 132 converts the extracted facial area into a face area based on a general image, and a more detailed description will be given.

As shown in FIG. 4, in the method of the present invention, an arbitrary object is simultaneously photographed by a general camera and an infrared camera, respectively. At this time, the face region is extracted from the image photographed by the infrared camera, and the extracted face region is converted into the face region of the general image based on the image photographed by the general camera, as shown in Fig. 5 . That is, in FIG. 5, the image of the infrared camera # 1 is converted into the image of the angle of view corresponding to the image of the general camera as in the image # 2, the image is converted into the image of moving the camera horizontally as shown in The resolution is converted into a converted image as shown in Fig. At this time, when the resolution is converted, the image is enlarged when the resolution of the general camera is higher than the resolution of the infrared camera, and in the opposite case, the image is reduced. FIG. 6 is a view showing a face region based on a converted general camera image in a resolution-converted image.

On the other hand, when the face region extraction is completed in step S204, the fake face determination unit 140 determines whether or not a thermal distribution exists in the extracted face region (step S206). That is, the fake face determination unit 140 determines whether or not there exists a thermal distribution silhouette or a map as seen in a general infrared ray photograph, for example, in the face area extracted by the face area extraction unit 131 Can be distinguished.

As a result of the determination in step S206, if there is no thermal distribution in the extracted face area, it is detected as a false face (step S207). If a thermal distribution exists in the extracted face area, it is detected as an actual face (step S209). That is, if there is no thermal distribution in the extracted face region, the forgery face determination unit 140 determines that the face is a face made of a photograph or the like rather than an actual face to be a forged face. If there is a heat distribution in the extracted face region, That is, it is judged to be a real face.

Preferably, the method further includes the step (S208) of generating a warning lamp or a warning sound indicating that the face is a forged face by the forged face determining unit 140 or the face authenticating unit 150 when the forged face is detected in step S207 can do.

Preferably, when the face is detected as an actual face in step S209, the facial feature size calculating unit 133 calculates the size of the facial feature with respect to the extracted face area using the face area conversion value for the general image (Step S210). Hereinafter, the facial feature size calculating unit 133 calculates the size of the face feature unit.

For example, when the angle of view of the general camera 111 and the angle of view of the infrared camera 112 are different from each other and are located on the same horizontal line at a certain distance, as shown in FIG. 7, The image of a specific size (face) located at a specific distance from the horizontal plane on which the image is located is captured at a specific position with a specific image size (number of pixels) in each camera. Accordingly, the facial feature size calculator 133 can determine the distance from the horizontal plane of the camera and the size of the object (face) by combining the image size and position of the object (face) photographed by each camera.

When the size of the facial feature size is calculated by the facial feature size calculator 133 as described above, the facial feature size data calculated by the facial feature size calculator 133 is preferably transmitted to the face authenticator 150), and comparing the received reference data with preset reference data and authenticating the actual face (S211). At this time, the face authentication unit 150 receives the facial feature size data calculated by the facial feature size calculator 133, for example, the received facial feature size data and the facial feature size If the comparison result is within the allowable error range as compared with the data, authentication can be performed by distinguishing the forged face from the actual face by a method of authenticating with an actual face.

As described above, the face authentication method according to the present invention searches for a face region in an existing general camera image (visible light image) by performing user authentication using a thermal image and a general image using an infrared camera and a general camera And it is possible to further enhance the reliability and security in face authentication.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments, but many variations and modifications may be made without departing from the spirit and scope of the invention. Be clear to the technician. Accordingly, the true scope of protection of the present invention should be construed according to the following claims, and all technical ideas within the scope of the same should be construed as being included in the scope of the present invention.

100 ... (invention) face authentication device 110 ... photographing part
111 ... ordinary camera 112 ... infrared camera
120 ... image acquiring unit 121 ... general image acquiring unit
122 ... thermal image acquisition unit 130 ... image processing unit
131 ... face area extracting unit 132 ... face area converting unit
133 ... facial feature size calculation unit 140 ... forged face judgment unit
150 ... face authentication unit

Claims (10)

a) capturing an arbitrary object by using a general camera and an infrared camera, respectively, to obtain a general image and a thermal image;
b) extracting a face region from the obtained thermal image by an image processing unit;
c) determining whether or not a thermal distribution exists in the extracted face area by the fake face determination unit; And
d) detecting as a false face if there is no thermal distribution in the extracted face region as a result of the discrimination; and detecting as a real face if a thermal distribution exists in the extracted face region.
The method according to claim 1,
And converting the extracted face region into a face region based on a general image based on the general image by using a face region converting unit using information on the extracted face region in the step b) Authentication method.
The method according to claim 1,
Further comprising the step of, when it is detected as a forged face in the step d), displaying a warning light or a warning sound indicating that the face is a forged face by the forged face determination unit or the face authentication unit.
3. The method of claim 2,
And calculating the size of the facial feature portion with respect to the extracted facial region using the facial region conversion value for the general image by the facial feature portion size calculating portion when the facial feature portion size is calculated in the step d) Face authentication method.
5. The method of claim 4,
Further comprising the step of authenticating the facial feature size data calculated by the facial feature size calculating unit with an actual face by receiving the facial feature size data calculated by the facial feature size calculating unit,
A photographing unit for photographing a subject in a visible light wavelength region and an infrared light wavelength region, respectively;
An image acquiring unit for acquiring a general image and a thermal image photographed by the photographing unit, respectively;
An image processing unit receiving the general image and the thermal image acquired by the image acquiring unit and extracting the face area, the face area conversion, and the size of the facial feature part;
A fake face determination unit for determining whether a face is forged based on a result of face region extraction from the image processing unit; And
And a face authentication unit that receives the processing results from the image processing unit and the fake face determination unit and authenticates the face.
The method according to claim 6,
Wherein the photographing unit includes a general camera for photographing a subject in a visible light wavelength region, and an infrared camera for photographing in an infrared wavelength region.
8. The method of claim 7,
The image acquiring unit includes a general image acquiring unit that acquires a general image photographed by a general camera of the photographing unit in a visible light wavelength range, a thermal image acquiring unit that acquires a thermal image photographed in an infrared wavelength region by the infrared camera of the photographing unit, And a face authentication unit.
9. The method of claim 8,
The image processing unit includes a face region extracting unit for extracting a face region from the thermal image received from the thermal image obtaining unit of the image obtaining unit, A facial region conversion unit for converting the facial region extracted by the facial region conversion unit into a face region based on the general image, and a facial feature size calculation unit for calculating the facial feature size from the face region based on the general image converted by the face region conversion unit And a face authentication unit.
10. The method of claim 9,
Wherein the forgery face determination unit determines whether or not a forged face is determined by determining whether or not a thermal distribution exists in the face region extracted by the face region extraction unit.

Images