KR20200107167A - Apparatus and Method for Making a Facial Image Suitable for Facial Recognition by Using Infraredat at Natural Lighting - Google Patents

Abstract

Disclosed are a face recognition device capable of acquiring an image capable of face recognition even in a natural light environment by using infrared rays, and a recognition method thereof. The face recognition apparatus of the present invention solves a conventional problem in which face authentication is impossible in an environment illuminated by natural light by acquiring a face image using infrared rays having a wavelength that is not conventionally used. The infrared face image acquired by the face recognition device of the present invention is suitable for face recognition even in daytime when natural light is a problem, and it is possible to generate an infrared face image by illuminating an infrared LED required at night when other lighting is insufficient.

Description

Facial Recognition Device and Method for Making a Facial Image Suitable for Facial Recognition by Using Infraredat at Natural Lighting}

The present invention relates to a face recognition apparatus capable of obtaining an image capable of face recognition even in a natural light environment by using infrared rays, and a recognition method thereof.

Face Recognition, like fingerprints and irises, is one of the things that is attracting attention as a personal authentication method, and recognizes a user by photographing a face with a camera and comparing it with a registered face image. Face recognition has an advantage that the user's body does not contact the device without requiring other manipulations of the user like iris recognition, and thus, face recognition has recently been incorporated in personal digital devices such as computers or mobile phones.

Face recognition technology is classified into a method using a 2D image and a method using a 3D shape. Since the 3D recognition technology requires a large amount of computation, a 2D face recognition system is generally used. On the other hand, if divided into the method of generating an image for face recognition, for face recognition, there is a method of photographing and recognizing a general black and white or color image using visible light, and a method of photographing and recognizing an infrared image using infrared rays. If necessary, both types of images may be used.

The method of using visible light such as black and white or color images has a limitation in that there must be a certain amount of illumination or more. Therefore, face recognition is impossible in environments where sufficient illumination is not provided, such as at night or in a dark room. In addition, there is a problem in that only a dark image is obtained when the subject face is backlit like a general camera.

Therefore, for face recognition, infrared images that can be used without sufficient illumination are mainly used. However, an infrared LED is required to illuminate the user with infrared rays. An infrared-type face recognition device illuminates a user's face with an infrared LED, and an embedded image sensor generates an infrared face image using an infrared image reflected from the user's face.

Since both the infrared method face recognition and the infrared method use light, they are inevitably affected by the external environment, and there are various problems depending on the external environment. One of them is that it is not possible to create an image that can recognize a face when external natural light is directly incident on the image sensor. For example, even if installed indoors as well as outdoors, if too much infrared rays are incident when a face recognition device is installed in a place affected by natural light, the brightness of most pixels is saturated and an infrared image that cannot even recognize the outline of the face is displayed. Can be created. If the infrared LED is turned off in the same situation, only the face area is darkened by backlighting.

In particular, since natural light incident through a window at sunset contains a lot of infrared rays, the user experiences faces in which faces are not recognized at work time when face recognition is concentrated. This is a key reason why face recognition devices are not widely used as professional personal authentication means such as access control.

Conventionally, the infrared region used for face recognition mainly uses infrared rays having a wavelength of approximately 850 nm among the near infrared (Near Infrared) bands. This is because most of the infrared cameras used to photograph users, such as security CCTV or interphone, have been using the 850nm band. It is also because the infrared filter of the 850nm band can be easily manufactured at a relatively very low price.

[Related invention]

1. A terminal capable of face recognition and its face recognition method (Korea Patent Publication No. 10-2009-0038110) This invention discloses a conventional face recognition method using a color image.

It is an object of the present invention to provide a face recognition device and a recognition method for obtaining an image capable of face recognition even in a natural light environment by using infrared rays.

The face recognition apparatus according to the present invention for achieving the above object includes a first LED unit, a first image sensor, and an infrared filter, and can generate the infrared face image even when exposed to natural light. Here, the first LED unit includes at least one infrared LED that emits infrared light having a wavelength of 30 nm to 950 nm and emits infrared rays toward a user who attempts to authenticate. The first image sensor finally generates an infrared face image of the user with light reflected and incident from the user's face. Meanwhile, the infrared filter transmits infrared rays having a wavelength of 930 nm to 950 nm in the light incident on the first image sensor.

According to an embodiment, the face recognition apparatus of the present invention includes: a second image sensor disposed in parallel with the first image sensor to generate a color image of the user's face; It may further include a second LED unit having at least one LED to emit light in a visible band for generating a color image by the second image sensor toward the user's face.

According to another embodiment, the face recognition device of the present invention further includes an illuminance sensor installed on an outer surface to detect illuminance, and a controller that controls the first and second image sensors to generate an image. The controller controls the first image sensor and the second image sensor to simultaneously generate an image in a first mode in which the illuminance detected by the illuminance sensor is greater than or equal to a reference value, and the first image sensor in a second mode in which the illuminance is less than the reference value. To create an image.

According to another embodiment, the face recognition apparatus of the present invention may further include a display unit that displays an image captured by the second image sensor to a user. The control unit may display a color image generated by the second image sensor in the first mode on the display unit, and display an infrared image generated by the first image sensor in the second mode on the display unit.

According to another embodiment, the first LED unit and the second LED unit are disposed around the first image sensor and the second image sensor to illuminate the center of the field of view of the first image sensor and the second image sensor. In this case, the face recognition device of the present invention may further include a third LED unit. The third LED unit is disposed above the first LED unit and the second LED unit, and includes at least one infrared LED and a visible light LED that emits at least one visible light. Illuminates the top of the image sensor's field of view.

A face recognition method of a face recognition device according to another embodiment of the present invention includes the steps of: at least one infrared LED emitting infrared light having a wavelength of 930 nm to 950 nm, emitting infrared rays toward a user attempting authentication; An infrared filter transmitting infrared rays having a wavelength of 930 nm to 950 nm in the light incident on the first image sensor; The first image sensor performs an infrared face image generation process including generating an infrared face image of the user with infrared rays reflected from the user's face and passed through the infrared filter, even when exposed to natural light The infrared face image may be generated.

The face recognition apparatus according to the present invention uses infrared rays having a wavelength of 940 nm rather than the conventionally widely used 850 nm wavelength band for face recognition. When the face recognition device of the present invention was installed and tested in an environment exposed to natural light, an infrared face image having sufficient contrast for face recognition was obtained, and the infrared ray was not saturated by natural light. Accordingly, the face recognition device of the present invention can generate an infrared face image to be used for face recognition even in an environment where external natural light is shining.

On the other hand, the face recognition device of the present invention can generate a color face image as well as an infrared face image, and the color face image is used as a user interface for displaying the face recognition process to the user, or when the user performs face authentication. It can be used as log data for authentication records to record facts.

1 is a view showing a face recognition device of the present invention,
2 is a block diagram of a face recognition device of the present invention;
3 is a flowchart provided to explain the operation of the face recognition device of the present invention, and
4 is a graph of characteristics of an infrared filter applied to the face recognition device of the present invention.

The present invention will be described in more detail below with reference to the drawings.

1 and 2, the face recognition apparatus 100 of the present invention includes a first LED unit 111, a second LED unit 113, a third LED unit 115, and a first camera unit 131. , A second camera unit 133, an illuminance sensor 151, a display unit 153, and a control unit 210, and includes a body 155 for accommodating them.

The face recognition device 100 of the present invention may be implemented as a dedicated device for face authentication, may be implemented in the form of an access control device for purposes such as access control, etc., or embedded in other devices such as a mobile phone or a computer. It can also be implemented in the form of Accordingly, the face recognition apparatus 100 may further include various configurations in addition to the configurations presented above as needed. For example, a user interface such as a keyboard (not shown) or other authentication means such as a fingerprint sensor (not shown), or a networking means such as a wireless LAN (not shown) may be included in the face recognition apparatus 100. .

The first LED unit 111, the second LED unit 113, the third LED unit 115, the first camera unit 131, and the second camera unit 133 detect the face of a user who wants to perform face recognition. It is disposed on the front surface 155a of the main body 155 so as to take pictures. The first LED unit 111 and the second LED unit 113 are disposed around the first camera unit 131 and the second camera unit 133 to provide a first camera unit 131 and a second camera unit 133. Provides lighting for shooting. The third LED unit 115 is disposed above the front surface 155a of the main body 155 and spaced apart from the first LED unit 111 and the second LED unit 113.

The first LED unit 111 includes at least one infrared LED as illumination for photographing the first camera unit 131 and emits infrared light that illuminates the user's face. The infrared rays used by the first LED part 111 are near infrared rays having a wavelength of 930 to 950 nm, and a wavelength of 940 nm is preferable. The sun's irradiance is highest in the visible region, and the longer the wavelength in the infrared region, the lower the illuminance. As described in the prior art, since the infrared wavelength used for photographing an infrared face image in the related art was 850 nm, it was more affected by natural light than at a wavelength of 900 nm or more. Therefore, in an environment in which natural light is directly or reflected, there is a problem that a facial contour cannot be recognized in an infrared face image photographed using an infrared ray of 850 nm band.

The applicant repeated the experiment to find an infrared wavelength suitable for image shooting in a band of 900 nm or higher, focusing on the fact that the natural light illuminance in the band of 900 nm or higher was lowered. It was confirmed that it is an infrared wavelength that can be obtained.

Infrared rays having a wavelength of 940 nm (±10 nm) emitted from the first LED unit 111 are reflected from the user's face and converted into an image of the user's face, and then re-enter the first camera unit 131.

The second LED unit 113 is provided with a plurality of general white light LEDs as illumination for the second camera unit 133, and emits white light in a visible light band. The white light emitted from the second LED unit 113 is reflected from the user's face, becomes an image of the user's face, and is incident on the second camera unit 133. The second LED unit 113 is disposed to surround the first camera unit 131 and the second camera unit 133 together with the first LED unit 111. When the first LED unit 111 includes a plurality of infrared LEDs and the second LED unit 113 includes a plurality of white LEDs, these LEDs are disposed in a mixed form.

The third LED unit 115 is not essential for photographing the first camera unit 131 and the second camera unit 133, but provides additional illumination for some users. The user's position (face position) suitable for capturing a face image is in front of the face recognition apparatus 100, and among them, is within the field of view of the first camera unit 131 and the second camera unit 133. The front surface 155a of the main body 155 is inclined at a certain angle, so that most users can take pictures with their head slightly bowed. In spite of the inclination of the front surface 155a of the main body, a tall user not only has to bend his back excessively, but also acquires a face image with a shadow on the face. The third LED part 115 is provided on the front side of the body 155a to illuminate the upper front side of the main body 155, that is, the top of the field of view of the first image sensor 235 and the second image sensor 255 do.

The third LED unit 115 may include at least one infrared LED and a visible light LED that emits at least one visible light.

The first camera unit 131 is a basic device for face recognition of the face recognition device 100 of the present invention, and includes an infrared filter 231, a first lens unit 233, and a first image sensor 235. Create an infrared image of the picture. The infrared image generated by the first camera unit 131 is provided to the controller 210.

As described below, the first camera unit 131 takes a picture under the control of the controller 210 in the standby mode and the recognition mode. During the standby mode, the shooting event occurs periodically, but in the recognition mode, there is one shooting event. When shooting is completed in the recognition mode, it is switched back to the standby mode.

In the standby mode, the first camera unit 131 periodically photographs the front in order to recognize the user's attempt to authenticate the face. In this case, the user's face may not be properly captured, or it is okay, so it is not necessary to capture the face area to an appropriate degree for face authentication. In addition, a plurality of images may be taken at each periodic shooting, but it is sufficient to take a single still image.

On the other hand, among the recognition modes performed for face recognition, the first camera unit 131 can capture a single infrared face image, but generates a plurality of infrared face images at a preset frame rate to increase the success rate of face authentication. It is desirable to do.

The infrared filter 231 passes an infrared band having a wavelength of 940 nm (± 10 nm) from the light incident to the first image sensor 235 and blocks light of the remaining wavelength. therefore. Visible light or infrared rays having a wavelength shorter or longer than 940 nm (± 10 nm) are blocked from being incident on the first image sensor 235.

As illustrated by way of example in FIG. 4, it is preferable that the infrared filter 231 is relatively clearly implemented in a band of a wavelength it passes through so that light of another wavelength is not incident. 4, the horizontal axis represents the wavelength and the vertical axis represents the transmission rate of the filter. The infrared filter 231 shown in FIG. 4 and applied to the present invention shows the performance of a band pass filter that passes a wavelength of 940 nm (± 10 nm).

Natural light has the highest illuminance of visible light and has considerable illuminance even in the 840 nm band, but when it approaches the 940 nm band, it is lowered to a level comparable to that of the first LED unit 111. Therefore, the face recognition apparatus 100 of the present invention using the 940nm band can exclude the influence of external natural light.

The first lens unit 233 forms an image of an infrared image reflected from the user's face and incident on the first image sensor 235 by adjusting the focal length and the angle of view so that the entire face of the user can be photographed. The first lens unit 233 may be provided between the infrared filter 231 and the first image sensor 235 or may be provided in front of the infrared filter 231. The first lens unit 233 generally uses one lens, but a combination of a plurality of lenses may be used.

Infrared rays having a wavelength of 940 nm (± 10 nm) emitted from the first LED unit 111 are reflected from the user's face, become an image of the user's face, and are incident on the first camera unit 131. In this case, the face image incident on the first camera unit 131 may include not only an image from external lighting or natural light, but also an image from light emitted from the second LED unit 113. The face image incident on the first camera unit 131 passes through the infrared filter 231 and becomes an infrared face image, and is then imaged on the first image sensor 235 by the first lens unit 233.

The first image sensor 235 generates a digital image, that is, an infrared face image, from the infrared face image that has passed through the infrared filter 231 and the first lens unit 233 and provides it to the controller 210.

The first image sensor 235 must include a wavelength of 930 to 950 nm in the operating bandwidth. The wavelength region of 930 to 950 nm need not be the center wavelength of the first image sensor 235. However, a general CMOS (Complementary Metal-Oxide-Semiconductor) image sensor is not only optimized for the visible light region, but also designed to significantly lower the quantum efficiency of the infrared region including the near infrared region to prevent image quality from deteriorating due to the influence of infrared rays. Therefore, it cannot be used as the first image sensor 235.

The second camera unit 133 includes a second lens unit 253 and a second image sensor 255 to capture a user's face and generate a color face image to be used for face recognition. Since the face recognition apparatus 100 of the present invention uses an infrared face image for face recognition, the second camera unit 133 is not for face recognition and is therefore not an essential component. The color face image generated by the second camera unit 133 is used as a user interface for displaying and guiding the face recognition process to the user, or an authentication recording log for recording the fact that the user has performed face authentication. ) Used as data. Accordingly, when entering the recognition mode described below, an image captured by the second camera unit 133 may be displayed through the display unit 153.

The second camera unit 133 operates under the control of the controller 210, and does not operate in the'second mode' among the operation modes of the face recognition apparatus 100 described below, and does not operate in the'standby mode'. May not.

The second camera unit 133 may have a conventional camera configuration, and any configuration known in the art as capable of generating a color image by photographing a subject may be used. The generation of the color face image by the second camera unit 133 is performed under the control of the controller 210, and the generated color face image is provided to the controller 210.

For example, the second lens unit 253 images a face image incident on the subject on the second image sensor 255. The second image sensor 255 is a conventional image sensor for generating a color image, and generates a digital color face image using a face image incident through the second lens unit 253.

The illuminance sensor 151 is mounted on the front surface 155a of the main body 155 and detects the illuminance in front of the main body 155 where a user's face to be recognized is located.

The display unit 153 is one of the user interface means, and displays various information or various control commands required for face authentication under the control of the control unit 210, or the user generated by the second camera unit 133 during the face recognition process. Display a color face image.

The control unit 210 performs the overall operation of the face recognition apparatus 100 of the present invention, and includes a mode control unit 211, a face detection unit 213, a feature point extraction unit 215, and a face recognition unit 217. Perform the face recognition process of the invention.

The mode control unit 211 controls an operation mode of the face recognition apparatus 100. The operation mode of the present invention includes a standby mode waiting for a user's face authentication attempt while checking the user's access degree, and a recognition mode performing an operation for face recognition according to the user's face authentication attempt.

The face recognition apparatus 100 generally operates in a standby mode, and when a user's approach is recognized, it switches to a recognition mode to perform a face recognition event. The conversion to the recognition mode occurs when the user's access is recognized. In order to recognize the user's access, a separate sensor or button operation may be used, or the first camera unit 131 or the second camera unit 133 may be used. May be.

For example, when a user operates a button or a proximity sensor confirms the user's access, it can be recognized as the user's contact. The button operation method is relatively less prone to other errors, but it is inconvenient because it requires a user's action, and the method using a sensor has a high possibility of malfunction. Alternatively, in the standby mode, the face detection unit 213 causes the first image sensor 235 or the second image sensor 255 to periodically photograph the front side to generate a face image. When a human face is detected in the generated face image, the detection information may be provided to the mode control unit 211. The mode control unit 211 may switch the standby mode to the recognition mode by using the detection information provided by the face detection unit 213.

Further, the recognition mode is divided into a first mode in which the illuminance value detected by the illuminance sensor 151 is equal to or greater than a preset reference value, and a second mode in which illuminance is less than the reference value. When switching from the standby mode to the recognition mode, the mode control unit 211 selects the first mode and the second mode based on the illuminance value detected by the illuminance sensor 151.

During the recognition mode, the face detection unit 213 controls the first image sensor 235 and the second image sensor 255 to generate a face image. In the first mode, the first image sensor 235 and the second image sensor Control 255 to simultaneously generate a face image, and in the second mode, only the first image sensor 235 is controlled to generate a face image.

The face detection unit 213 detects a face area from an image provided by the first camera unit 131.

First, in the recognition mode, the face detection unit 213 performs a pre-processing process for face recognition. A face provided by the first camera unit 131 at a preset frame rate while the first LED unit 111 is turned on. Extract the human face from the image. A knowledge-based learning method, a feature-based method, a template matching method, an appearance-based method, and the like may be used to extract the face part, and any conventionally known method may be applied.

The appearance-based method is, for example, by applying a conventional image processing method to an infrared face image provided by the first camera unit 131 to extract an outline of a face corresponding to a roughly circular shape, and then extract the eyes and mouth based on the location from the face outline. By detecting, the face contour is recognized.

The operation of the face detection unit 213 in the standby mode is used to confirm the user's authentication attempt. During the standby mode, the face detection unit 213 controls the first camera unit 131 to periodically photograph the front side while the first LED unit 111 is turned off. When the first camera unit 131 provides an image photographed in front, the face detection unit 213 may check whether or not the user attempts to authenticate by checking the degree to which a person's face is present in the image. The image processing at this time does not need to be a preprocessing process for extracting feature points from the face image. Since many images are not required to check whether the user attempts to perform face authentication, it is sufficient for the first camera unit 131 to take a still image or generate an image at a frame rate much lower than the frame rate for face recognition. .

The feature point extracting unit 215 extracts feature points of the user's face from the face area detected by the face detecting unit 213 for face recognition. The extracted feature points may be used for user registration or may be used as an authentication means for comparing with a previously registered user.

The feature points extracted by the feature point extraction unit 215 are features that reflect the shape information of the face peculiar to the user, such as information on the shape, length, and relative position of the face, chin, eyes, eyebrows, nose, mouth, etc. Included. In addition, the feature points of the face can be extracted by applying various information and extraction methods known in the art. However, the feature point extracting unit 215 of the present invention is characterized in that it uses an infrared face image.

The face recognition unit 217 authenticates a user who attempts face authentication by comparing the feature point data extracted by the feature point extracting unit 215 with previously registered'user registration information'. The user registration information may include feature point data extracted from a user's face image, a user image (color and/or infrared image), information about a user (name, etc.). The face recognition apparatus 100 may itself perform a user registration process to hold user registration information, or may receive and hold user registration information from another authentication device or server.

Face recognition device operation

Hereinafter, the operation of the face recognition apparatus 100 of the present invention will be described centering on the operation of the control unit 210. The basic face recognition method of the present invention includes the step of confirming the user's attempt to authenticate the face, and the face detection unit 213 extracts a face portion for each of a plurality of face images generated by the first camera unit 131 at a preset frame rate. , The feature point extraction unit 215 is performed by extracting facial feature point data for each of a plurality of face images.

<Confirmation of face authentication attempt: S301, S303>

The face recognition device 100 basically operates in the standby mode. During the standby mode, the face detection unit 213 causes the first camera unit 131 to periodically photograph the front side to generate an image and provide it to the control unit 210. At this time, the first LED unit 111 may be turned on, but it is preferable to take a picture of the user in external lighting or natural light by turning off the first LED unit 111 (S301).

The face detection unit 213 detects a human face by performing image processing on the image captured by the first camera unit 131. When the face detection unit 213 recognizes a person's face, the mode control unit 211 determines that there is a face authentication attempt. The face detection unit 213 may perform a degree of detecting the outer face of the face by detecting an edge in the face image (S303).

As described above, the face recognition device 100 further includes a proximity sensor or a button, and the control unit 210 is an authentication initiating unit (not shown) that can confirm a face authentication attempt based on the proximity sensor or the user's button manipulation. It may further include.

<Selection of operation mode for face authentication: S305, S307>

When the user's face is recognized in step S303, the mode control unit 211 switches the operation mode from the standby mode to the recognition mode, and checks the illuminance detected by the illuminance sensor 151 (S305).

If the illuminance is higher than the reference value, the first mode is switched, and if the illuminance is less than the reference value, the second mode is switched (S307).

<Create face image according to recognition mode: S309, S311>

According to the mode change of the mode control unit 211, the face detection unit 213 controls the first camera unit 131 to generate a plurality of infrared face images at a preset frame rate. The face detection unit 213 turns on the first LED unit 111 and controls the first camera unit 131 to photograph the user at a preset frame rate to generate an infrared face image.

Meanwhile, when the operation mode is the first mode, the face detection unit 213 controls the second camera unit 133 to generate a plurality of color face images at a preset frame rate at the same time as the first camera unit 131. Accordingly, the face detection unit 213 turns on the second LED unit 113 and controls the second camera unit 133 to photograph the user at a preset frame rate to generate a color face image. Meanwhile, during the first mode, the color face image generated by the second camera unit 133 is displayed through the display unit 153 (S309).

When the operation mode is the second mode, the face detection unit 213 does not proceed to photograph the second camera unit 133. An infrared face image captured by the first camera unit 131 is output on the display unit 153 (S311).

<Face Recognition and Authentication: S313>

The face detection unit 213 recognizes a face part from the infrared face image captured by the first camera unit 131, and the feature point extraction unit 215 extracts feature points of the face (S309). The extracted feature points are used for user registration as necessary or used for user authentication by comparing with previously registered user information.

The face recognition apparatus 100 of the present invention operates in the above manner.

In the above, preferred embodiments of the present invention have been illustrated and described, but the present invention is not limited to the specific embodiments described above, and is generally used in the technical field to which the present invention belongs without departing from the gist of the present invention claimed in the claims. Various modifications are possible by those skilled in the art of course, and these modifications should not be individually understood from the technical idea or perspective of the present invention.

Claims (10)

In the face recognition device,
A first LED unit including at least one infrared LED emitting infrared light having a wavelength of 930 nm to 950 nm and emitting infrared rays toward a user attempting authentication;
A first image sensor generating an infrared face image of the user with light reflected from the user's face and incident; And
A face recognition device comprising an infrared filter that transmits infrared rays having a wavelength of 930 nm to 950 nm from light incident to the first image sensor, and generates the infrared face image even when exposed to natural light.
The method of claim 1,
A second image sensor disposed in parallel with the first image sensor to generate a color image of the user's face; And
A face recognition apparatus comprising: a second LED unit including at least one LED to emit light in a visible band for generating a color image by the second image sensor toward the user's face.
The method of claim 2,
An illuminance sensor installed on the outer surface to detect illuminance; And
The first image sensor and the second image sensor are controlled to generate an image, but in the first mode in which the illuminance detected by the illuminance sensor is equal to or greater than a reference value, the first image sensor and the second image sensor are controlled to simultaneously generate an image. And a controller configured to control the first image sensor to generate an image in a second mode in which the illuminance is less than the reference value.
The method of claim 3,
Further comprising a display unit for displaying the image captured by the second image sensor to the user,
Wherein the control unit displays a color image generated by the second image sensor on the display unit in the first mode, and displays an infrared image generated by the first image sensor on the display unit in the second mode Face recognition device.
The method of claim 2,
An angle of view of the first image sensor and the second image sensor by having at least one infrared LED and a visible light LED that is disposed above the first LED unit and spaced apart from the second LED unit and emits at least one visible light Further comprising a third LED to illuminate the upper portion of the,
The first LED unit and the second LED unit are provided around the first and second image sensors to illuminate a center of a field of view of the first and second image sensors.
In the face recognition method of the face recognition device,
At least one infrared LED emitting infrared light having a wavelength of 930 nm to 950 nm emitting infrared rays toward a user attempting authentication;
Transmitting infrared rays having a wavelength of 930 nm to 950 nm in the light incident on the first image sensor by the infrared filter; And
The first image sensor performs an infrared face image generation process including generating an infrared face image of the user with infrared rays reflected from the user's face and passed through the infrared filter, even when exposed to natural light Face recognition method of a face recognition device, characterized in that capable of generating the infrared face image.
The method of claim 6,
During the process of generating the infrared face image,
At least one LED emitting light in a visible ray band toward the user's face to generate a color image by the second image sensor; And
A face recognition method of a face recognition apparatus, characterized in that a second image sensor is disposed in parallel with the first image sensor to perform a color face image generation process including generating a color image of the user's face.
The method of claim 7,
Detecting illuminance using an illuminance sensor installed on the outer surface of the main body; And
The control unit further comprises controlling the first image sensor and the second image sensor to generate an image,
The controller performs both the infrared face image generation process and the color face image generation process in a first mode in which the illuminance detected by the illuminance sensor is greater than or equal to the reference value, and the color face image generation process in the second mode when the illuminance is less than the reference value Face recognition method of a face recognition device, characterized in that not performing.
The method of claim 8,
In the first mode, an image captured by the second image sensor is displayed to the user through a display unit, and in the second mode, an infrared image generated by the first image sensor is displayed on the display unit. Device face recognition method.
The method of claim 7,
At least one infrared LED disposed at an upper portion spaced apart from the first LED unit and the second LED unit, and a visible light LED emitting at least one visible light are positioned above the field of view of the first and second image sensors. Further comprising the step of lighting,
The first LED unit and the second LED unit are provided around the first image sensor and the second image sensor to illuminate the center of the field of view of the first image sensor and the second image sensor. Way.

Images