KR20110006062A - System and method for face recognition - Google Patents

Abstract

PURPOSE: A face recognition system and a method thereof for accurately identifying a face within a photographing area are provided to minimize the probability of failure of the face recognition system. CONSTITUTION: An image pickup unit(20) takes a photograph of a user to attempt face recognition and generates a user image. A face recognition processing unit(30) determines if there is a match with a registered face stored in a database. The face recognition process unit implements face recognition using a predetermined facial feature. A display unit(40) visually displays a part of the face.

Description

Face Recognition System and Face Recognition Method {SYSTEM AND METHOD FOR FACE RECOGNITION}

The present invention relates to a face recognition system and a face recognition method, and more particularly, to a face recognition system and a face recognition method capable of inducing a user to photograph a user's face in a recognizable size and shape in face recognition. It is about.

Recently, in order to block and allow access to various computing systems requiring security management or access to areas under security management, various face recognition systems for recognizing a user's face and performing authentication processing have been variously developed and applied to the corresponding fields. It is becoming.

In such a face recognition system, by using a camera to photograph an authenticated user, a registered image, which is an image for user authentication, is generated in advance, a face shape is detected from the generated registered image, and each face part is detected from the detected face shape. Is detected, a feature point is designated at each detected face part, and the specified feature point is digitized and stored as feature point information.

When the user's access to the face recognition system occurs and a request for user authentication occurs, the face recognition system photographs the user's body, detects the user's face from the captured image, and detects each of the user's faces. Detect a face part, designate a feature point in the detected face part, digitize the specified feature point, compare the information on the digitized feature point with the feature point information generated for the pre-stored registered face, and compare the degree of agreement with the result of the comparison. Score or percentage and compare the calculated value with a predetermined reference range to determine the success or failure of authentication.

However, such a face recognition system generates a user image by photographing a user at the time of input of a user's face photographing command or when the user's body is detected within a predetermined photographing area or a human body detection area. do.

At this time, if the user is located too far from the camera or is positioned too far from the center of the camera photographing area, it is difficult for the captured user's face to be accurately captured sufficiently to perform face recognition.

1 is a diagram illustrating an example of a user image photographed by a face recognition system. Figure 1 (a) shows a preferred image taken with the user's face accurately positioned at the right size in the center of the shooting area. FIG. 1 (b) shows a state where the user is positioned too far away from the center of the shooting area, and FIG. 1 (c) shows a state where the user is located too far from the camera and cannot accurately recognize the user's face. have.

Among these, if face recognition is attempted using the images as shown in FIGS. 1B and 1C, the face recognition system may not accurately analyze the user's face and may consider the face recognition failure. Therefore, if authentication failure is determined despite being a legitimate user, the user will feel inconvenience in face recognition processing.

An object of the present invention for solving the above-mentioned problems is to induce a user to accurately position his or her face in a photographing area of a face recognition system, thereby minimizing the probability of failure of face recognition.

It is another object of the present invention to minimize the case where face recognition fails by the user being placed too far from the camera.

The present invention for achieving the above object, the database unit for storing one or a plurality of faces as a registered face; A photographing unit which photographs a user who attempts to recognize a face and generates a user image; A face recognition processor configured to perform face recognition by determining whether a user face detected in the generated user image matches the registered face stored in the database unit; The face recognition processing unit may perform face recognition using the user image only when a predetermined face part of the detected user face is located within a preset restriction area.

The display apparatus may further include a display unit for visually displaying the user image, wherein the display unit visually displays at least a portion of the restriction area.

In addition, a part of the restricted area is displayed on the surface of the display unit by printing or stamping.

The display unit may invert and display the user image.

In addition, the predetermined face part may be both eye parts, and the face recognition may be performed only when both eye parts are positioned within an eye position limit region preset in the display unit.

The face recognition may be performed only when the distance between the two eyes is within a predetermined limit distance range.

In addition, the eye position limiting area is divided into left and right and set independently, the left eye of the user face is located within the left eye position limiting area, and the right eye is located only within the right eye position limiting area. It is characterized by performing.

The size of the left eye position limiting region or the right eye position limiting region may be arbitrarily set and changed.

The distance between the left eye position limiting region and the right eye position limiting region may be arbitrarily set and changed.

The predetermined face part may be a face outline detected from the user's face, and face recognition may be performed only when at least a part of the line shapes constituting the face outline are located within a preset face outside restriction area. do.

In addition, the size or shape of the face outside restriction area may be arbitrarily set and changed.

In addition, the face outer limit area is made of a face area minimum limit border and a face area maximum limit border, and the size or shape of the face area minimum limit border and the face area maximum limit border may be arbitrarily set and changed. It is characterized by.

In addition, the predetermined face part is a face outline for both eyes and the user's face, the both eye parts are located within a preset eye position limit region, and at least a part of the line shapes constituting the face outline are preset. It is characterized in that the face recognition is performed only when the position is located within the limited area outside the face.

In addition, the size or shape of the eye position limiting region or the face outer limiting region may be arbitrarily set and changed.

The apparatus may further include a human body approach detecting unit that detects the human body's approach to the front of the camera unit, and when the human body access detecting unit detects a user's approach to face recognition, maintains a wake-up mode to recognize the face. In the case where the user is not detected, power consumption may be minimized by maintaining the sleep mode.

Another aspect of the present invention for achieving the above object comprises the steps of photographing a user attempting a face recognition to generate a user image; Detecting both eye parts of the user from the generated user image; Determining whether the detected both eye parts are located within a preset eye position limit area; When both eye parts are located within the eye position limiting area, face recognition is performed by determining whether a user face detected from the user image matches a registered face registered in advance, and the eye parts are located in the eye position. If not located within the restricted area, instructing the user to change the position of the face; It provides a face recognition method comprising a.

In addition, when both eyes are located within the eye position limiting area, the face is further included when at least some of the line shapes constituting the face outline detected from the user face are located within the preset face limiting area. Characterized in that the recognition.

Another aspect of the present invention for achieving the above object comprises the steps of photographing a user attempting a face recognition to generate a user image; Detecting a shape of a line forming an outline of a user's face in the generated user image; Determining whether at least a part of the detected line shape is located within a preset face outer limit area; If at least a portion of the line shape is located within the face outside restriction area, face recognition is performed by determining whether a user face detected from the user image matches a registered face registered in advance, otherwise face recognition is performed. Inducing the user to change the position of the face without performing; It provides a face recognition method comprising a.

According to the present invention as described above, by inducing the user to accurately position his face in the photographing area of the face recognition system, by accurately photographing the user's face to clearly extract the feature points of each part of the face probable failure of face recognition Can be minimized.

That is, in the present invention, even though the user is located too far away from the camera or positioned too far from the center of the shooting area, the user image for face recognition is acquired, and the face is misrecognized even though the user is already registered. This can be prevented.

In addition, the present invention, by repeatedly performing the face recognition procedure by the misrecognition, it is possible to minimize the waste of power due to the long-term operation of the system. In other words, the user image for accurate face recognition can be obtained quickly, thereby minimizing the probability of misrecognition of the registered user, thereby minimizing the driving time of the face recognition system. Minimizing the running time of the system means minimizing the power consumption of the system.

In addition, the present invention can significantly reduce the overall power consumption by maintaining the system in a low power consumption state when no user attempts to recognize the face.

Hereinafter, the detailed configuration of the present invention will be described with reference to the drawings.

2 is a block diagram briefly illustrating a configuration of a face recognition system according to an exemplary embodiment of the present invention. Referring to FIG. 2, the face recognition system includes a database unit 10, a photographing unit 20, a face recognition processing unit 30, and a display unit 40.

The database unit 10 is configured to store a user image generated by photographing an authenticated user as a registered image, and to read it arbitrarily if necessary. At this time, the face image generated by detecting only the user's face from the registered image as well as the registered image, the coordinate information of the feature points set for each face region from the face image, or the coordinates of the set feature points are digitized by a function such as a vector. The expressed information (feature point information) can also be retained.

One or more authenticated users may be registered in the database unit 10, and one or more registered images and / or feature point information calculated from each registered image may be maintained for each authenticated user.

The photographing unit 20 includes a camera 22 having a predetermined photographing area, and photographs a photographing area as a moving picture or a still image according to a photographing signal input from an external device (for example, a face recognition processing unit). Digital images are generated and output. In this case, the photographing unit 20 preferably photographs and outputs a continuous still image, whereby the video is shown to be photographed.

In addition, the photographing unit 20 may be provided with lighting means for illuminating the front by irradiating infrared or visible light (not shown).

The face recognition processing unit 30 enters a photographing area (see 'P' in FIG. 3) of the photographing unit 20 and generates a digital image generated by photographing a user requesting to perform face recognition (for user image or face recognition). Calculating the feature point information from the user image, and comparing the feature point information calculated from the user image with the feature point information of one or a plurality of authenticated users stored in the database unit 10 to calculate the degree of coincidence (matching degree). The recognition process is performed.

At this time, the face recognition processing unit 30 performs the face recognition processing only when the user's face is photographed with an appropriate size and resolution, so that the face by the unsuitable image can not clearly detect the feature points on the user's face As the recognition is performed, it is desirable to minimize the case where the user's face is misrecognized and determined to be an authentication failure.

In addition, the face recognition processing unit 30 continuously analyzes the user image output by the photographing unit 20 to track the position of the user's face (or a specific area of the face, more preferably the eye area), and It is preferable to generate a user image for face recognition to be applied to face recognition by generating a still image by capturing a moving image at a time when a face is photographed with a size suitable for a suitable position in the photographing area. As such, when the user image for face recognition is automatically generated by eye position tracking, even when the user approaches the camera 22 and moves constantly, the user image can be automatically generated at the most suitable moment, so that the face recognition is performed. In the present invention, an image suitable for face recognition can be obtained quickly even without artificially inducing a shooting to a user.

Of course, when the face is located at the most suitable position while the user approaches the camera 22, the user may induce the movement to stop for a while, or may guide the user to press a button for taking a picture. .

The display unit 40 visually displays the digital image photographed by the camera 22 of the photographing unit 20. In this case, the display unit 40 preferably outputs the inverted left and right of the photographed digital image.

In the conventional face recognition system, the photographed user image is displayed as it is being photographed. In this case, the user in the displayed image moves opposite to the motion of the actual user. Therefore, when the user views the displayed image and wants to move the position to the center of the image, the user is shown to move as opposed to the actual movement, which may cause great confusion to the user.

Therefore, in the display unit 40 of the face recognition system according to the present invention, by inverting and displaying the displayed user image, the user can easily adjust the position of the face while watching his / her appearance displayed as a general mirror. do.

Meanwhile, in the above description, the operation of the face recognition processing unit that performs face recognition processing will be described in more detail with reference to FIGS. 3 to 5.

FIG. 3 is a diagram for describing an eye position limiting area applied to a face recognition system according to an exemplary embodiment of the present invention, which is applied to allow a user's face to be accurately captured. In FIG. 3, a rectangular region of the photographing area P, a vertical line and a horizontal line for indicating the center of the photographing area, and an oval shape centering on the photographing area P, the face area border for guiding a preferable position of the user's face. (F) and an example of the eye position restriction area E for guiding a preferable position with respect to both eye parts of a user's face as a rectangular shape are shown.

Each of these areas may be hidden from the user in the display unit 40, but at least a portion (eg, a face area border) may be visually displayed for viewing by the user.

In FIG. 3, the face region edge F is illustrated in an ellipse shape, but it is also possible to apply the face region edge F in a rectangular shape or other various forms as well as an ellipse shape.

In general, face recognition first detects both eyes of a user's face. This is because detecting the eye area among the face parts is the simplest and most precise location can be specified, which can be a reference for detecting the user's face.

When both eye parts are detected in this manner, it is determined whether the detected eye parts are located within the eye position limiting area E preset in the photographing area. At this time, the eye position restriction area E is set slightly upward from the center of the face area border F set in the photographing area P. FIG.

When both eye parts of the user's face are located within the eye position limiting area E, the current user image is processed by the user's face being positioned near the center of the shooting area (for example, the face). It is determined that the image is a suitable image that can be suitably used for the recognition process, the eye gap inspection process, or the face outline inspection process).

On the other hand, if both eyes of the user's face is not located within the eye position restriction area, the display unit 40, such as "Please move to the center of the screen," "Please put your face in the frame", etc. After the guidance is performed visually or acoustically, the user image is acquired and analyzed again.

On the other hand, since the image to be photographed is shown inverted left and right on the display unit 40, the user can adjust the position of the face while looking at his image displayed in the same way as the mirror.

4 is a diagram illustrating a state in which both eye parts of the user's face are located within an eye position limit region. In FIG. 4, in the state where only the eye position restriction region E is displayed on the display unit 40 of the various regions of FIG. 3, both eye portions of the user's face are correctly positioned within the eye position limitation region E. In FIG. Show (fit image).

On the other hand, in Fig. 5, not only both eye parts of the user's face are located in the eye position limiting area E, but also the eye position limiting area E is bisected, the left eye is left eye position limiting area EL. Located inside, and the right eye is located within the right eye position restriction area ER. When the respective eye position restriction region is applied to each eye as described above, an image of a user face more suitable for face recognition can be obtained.

In addition, FIG. 5 illustrates a process of measuring the distance D between both eyes detected from the user's face.

When the distance D between both eyes is measured in this way, it is determined whether the measured distance D falls within a predetermined reference range. Subsequently, subsequent processing such as face recognition can be executed only when the distance D is within the reference range.

If the distance D falls below or exceeds the reference range, it is considered that the user is too close to the camera 22 or is too far away, and “get a little closer”, “a little further away” Etc. "can be performed visually or audibly.

Such a measurement procedure of the distance D may be performed after it is determined that both eyes are located in the eye position limitation area E as shown in FIG. 4, and each eye position limitation area EL, as shown in FIG. 5. ER), after each eye is determined to be located.

Of course, it can be done when any one eye is located in any one of each eye position limiting area (E, EL, ER), or even when no one eye is located, the eye position limiting area (E) Independent of the determination using EL, ER, the determination of a suitable image may be performed by measuring only the distance D. FIG.

In addition, in the face recognition system according to an embodiment of the present invention, the eye position limiting area E for determining whether both eye parts are located inside, the left eye position limiting area EL for each eye part, and Preferably, the horizontal and / or vertical size of the right eye position restriction area ER can be set arbitrarily. Further, it is also possible to independently set the sizes of the left eye position limiting area EL and the right eye position limiting area ER, and the distance between the left eye position limiting area EL and the right eye position limiting area ER. It is also possible to set and change arbitrarily.

In adjusting the area as described above, if the size of each eye position limiting area (E, EL, ER) is set small, the user may feel somewhat uncomfortable to accurately position his / her eyes, but for face recognition acquired The user image is more likely to be a suitable image having a desired size and resolution, and more accurate facial feature point information can be obtained, thereby improving security.

On the contrary, if the size of each eye position limiting area (E, EL, ER) is set large, the user can easily position the eye area in the corresponding area, which improves convenience, but accurately and clearly displays the feature point information on the user's face. There is a disadvantage that the security is lowered due to the high possibility of not being detected.

In addition, in the face recognition system according to the exemplary embodiment of the present invention, the distance D between both eye parts of the user located in the eye position limiting area E (or both eye position limiting areas EL and ER) is shown. It is desirable to be able to arbitrarily set and change the reference range for.

That is, if the reference range for the distance D is set to be large, the user will have to bring the face closer to the camera 22, thereby increasing the inconvenience of face recognition. Feature points of the face region can be clearly detected so that face recognition can be made more accurately. On the contrary, if the reference range is set small, the user image for face recognition can be obtained even at a position far away from the camera 22, thereby improving user convenience, but since the user's face is taken small, There is a drawback that the probability of face recognition failure increases because the likelihood that the feature points are not detected accurately increases.

As described above, by using various eye position limiting areas (E, EL, ER) and the distance D between both eyes, a user's face image close to the center of the screen and a suitable size applicable to face recognition can be obtained. In this way, it is possible to minimize the possibility of authentication failure due to misrecognition caused by face recognition using an incorrectly photographed image.

Next, another method for obtaining a suitable image will be described with reference to FIG. 6. FIG. 6 illustrates a state in which a range of the face outline restriction area including a face area minimum restriction border F1 and a face area maximum restriction border F2 having a function similar to that of the face area border F shown in FIG. 3 is set.

In order to apply the range of the face outline restriction area, first, the face recognition processor 30 should detect the user's face in the user image and detect the outline from the detected face. That is, both eyes of the user's face area are detected, and each face area (for example, eyebrow, nose, mouth, ear, chin, cheek, etc.) is detected based on both detected eyes, and the detected face area is detected. The shape of the line forming the outline of the face is detected as the face outline. Then, it is checked whether the detected face outline is located at least in the range between the face area minimum limiting border F1 and the face area maximum limiting border F2.

At this time, when the face outline is located between the face area minimum limit border F1 and the face area maximum limit border F2, subsequent processing (for example, face recognition) is performed, and the face outline is limited to the face area minimum limit. When located in the inner region of the border (F1) or in the outer region of the face area maximum limiting border (F2) to guide the user to get closer to the camera 22 or to move away.

On the other hand, a check as to whether the detected face outline is located in the range between the face area minimum limiting border F1 and the face area maximum limiting border F2 is one of the line shapes forming the outline of the user's face in the face recognition processing unit. It is made in such a way as to determine whether or not a line shape of a predetermined amount or more is within the range.

More specifically, for example, at least 70% of the line shape of the face outline is located outside the face area minimum limiting border F1 and at least 70% of the line shape is of the face area maximum limiting frame F2. In the case where it is determined to be located in the inner region, it may be determined that the face outline is located in a range between the face region minimum limiting border F1 and the face area maximum limiting border F2.

In such a configuration, when only one of the face area minimum limiting border F1 or the face area maximum limiting border F2 is displayed on the display unit 40, the user may roughly display his or her face at a given edge. By matching, it is possible to obtain a suitable image.

In fact, if both the face area minimum limit border F1 and the face area maximum limit border F2 are displayed on the display unit 40, and the user wants to place the outline of his face within the range, this is inconvenient for the user. Will result. Therefore, it is preferable to display only the face area maximum limiting border F2 or to display only a border (not shown) indicating an intermediate region between the face area minimum limiting border F1 and the face area maximum limiting border F2.

In the present invention, by applying such a range of the facial outer limit region, it is possible to obtain a suitable image in which the user's face is photographed at an optimal size.

In addition, the facial recognition system according to an exemplary embodiment of the present invention may further include a human body approach detecting unit (not shown) for detecting a user's approach to face recognition. For example, the human body approach detection unit is configured as a pair of light emitting elements / light receiving elements to detect the light reflected by the human body to detect the approach of the human body, or to detect the energy emitted from the human body such as an infrared detection sensor to approach the human body. Can be detected.

As described above, the face recognition system using the human body approach detecting unit is kept in a sleep mode to minimize power consumption in the usual state (when there is no user's approach to face recognition), and the face is detected by the human body approach detecting unit. When the user's approach to the recognition is detected, the user wakes up to generate a user's image and performs a subsequent face recognition procedure.

According to such a configuration, it is possible to minimize the operating time of components such as the camera 22 with a large power consumption in the face recognition system, thereby minimizing the overall power consumption.

Next, an example of a face recognition method by the face recognition system according to the present invention will be described with reference to FIG. 7.

First, although not shown, a procedure of storing a registered image in the database unit 10 will be described. The registered image may be made by an operator. After accurately positioning the authenticated user's face in the photographing area of the camera 22, the operator sends a command to instruct the user to directly photograph the user's face. Generates a registered image by shooting and obtaining a still image for the image. The generated registered image is stored in the database unit 10 together with information such as identification information of the authenticated user or creation date of the registered image. In addition, the feature point information on the feature points of each part constituting the user's face in the registered image may be stored at the same time.

In addition, only one registered image (and / or feature point information) stored in this manner may be registered for one user, and a plurality of registered images may be registered. In addition, in some cases, it is also possible to automatically update the registered video stored with the new registered video. That is, after a predetermined time has elapsed or at the request of an authenticated user (or operator), the registered image which has already been stored is updated to the latest registered image, thereby changing the face of the authenticated user (eg, aging, illness or Change due to molding, etc.).

Next, referring to the face recognition method of FIG. 7, the face recognition system according to the present invention minimizes power consumption by maintaining a sleep mode. In this state, when the user approaches the facial recognition system in order to attempt to recognize the face, the human body access detection unit (which is configured to detect the human body's approach periodically and continue to operate in the sleep mode) wakes up the facial recognition system. In operation S12, the wake-up face recognition system displays a user image generated by continuously photographing a user approached through the photographing unit 20 in real time (S12).

When the user image is captured and generated, the face recognition processing unit 30 of the face recognition system detects a face part of the user, and detects a specific face part (for example, both eyes) from the detected user face (S14). .

When both eye parts are detected in this way, it is determined whether the detected eye position is inside the preset eye position limit area (S16, S17). If it is determined that the eye part is located inside the eye position limiting area E, the process proceeds to step S18, and if the eye part is not within the eye position limiting area, the user is guided to move the face position (S26). ).

On the other hand, when it is determined that the eye part is inside the eye position restriction area E, the outline of the user's face is subsequently detected (S18).

When the outline of the user's face is detected, it is determined whether the detected outline is located within the preset face outline restriction area (S20, S21). In this case, the determination may be performed by calculating a ratio of the outline existing in the face limitation region among the entire portions of the user's face outline and comparing it with the reference ratio. For example, when 70% of the user's face outline exists within the face outline restriction area, it may be determined that the outline of the face detected from the user's face is within the face outline restriction range.

On the other hand, if it is determined by this determination that the outside of the user's face is located inside the face outside limitation area, the current user image is set as the face recognition user image and then the user's face included in the face recognition user image. The face recognition is performed by using S22. On the contrary, if it is determined that the outside of the user's face does not exist in the face outside restriction area, the process proceeds to step S26 to guide the user.

Meanwhile, in the present invention as described above, each eye position limiting area (E, EL, ER), face area border (F), face area minimum limiting border (F1), and face area maximum limiting border (F2) The display method may be arbitrarily set by a user or an operator, and a display method such as a shape or color of a border line may be displayed on the display 40 and displayed together with the generated user image.

However, alternatively, it is also possible to display various borders directly on the surface of the display unit 40. That is, it is also possible to apply a method such as drawing a border line using a material having a color of itself such as a fluorescent material on the panel surface of the display unit 40.

If the border line is directly formed in this manner, a software processing process for displaying the border line of each area on the display unit 40 can be omitted, thereby enabling a faster face recognition process.

Although the above-described face recognition method has been described as simultaneously using the eye position limiting region and the face outer limiting region, another embodiment using only the eye position limiting region or only the face outer limiting region is possible.

The foregoing description is merely illustrative of the technical idea of the present invention, and various changes and modifications may be made by those skilled in the art without departing from the essential characteristics of the present invention. Therefore, the embodiments disclosed in the present invention are not intended to limit the technical spirit of the present invention but to describe the present invention, and the scope of the technical spirit of the present invention is not limited by these embodiments. The protection scope of the present invention should be interpreted by the following claims, and all technical ideas within the equivalent scope should be interpreted as being included in the scope of the present invention.

1 is a diagram illustrating an example of a user image photographed by a face recognition system according to an exemplary embodiment of the present invention.

2 is a block diagram briefly illustrating a configuration of a face recognition system according to an exemplary embodiment of the present invention.

3 is a view for explaining an eye position limiting region and a face region border in a face recognition system according to an exemplary embodiment of the present invention.

4 is a diagram illustrating a state in which both eye parts of the user's face are located within an eye position limit region.

FIG. 5 is a view illustrating a state in which a left eye is located within a left eye position restriction area and a right eye is located in a right eye position restriction area of the user's face.

Fig. 6 is a diagram illustrating a range of the face outside restriction area.

7 is a flowchart illustrating a face recognition method according to an embodiment of the present invention.

Claims (18)

A database unit for storing one or a plurality of faces as registered faces; A photographing unit which photographs a user who attempts to recognize a face and generates a user image; A face recognition processor configured to perform face recognition by determining whether a user face detected in the generated user image matches the registered face stored in the database unit; Including, And wherein the face recognition processing unit performs face recognition using the user image only when a predetermined face part of the detected user face is located within a preset restriction area. The method of claim 1, Further comprising a display unit for visually displaying the user image, And the display unit visually displays at least a portion of the restricted area. The method of claim 2, And displaying part of the restricted area on the surface of the display by printing or engraving. The method of claim 2, And the display unit inverts and displays the user image. The method according to any one of claims 2 to 4, The predetermined face part is both eye parts, And face recognition only when the both eye parts are positioned within the eye position limiting region preset in the display unit. The method of claim 5, And face recognition only when the distance between the two eyes is within a predetermined limit distance range. The method of claim 5, The eye position limiting area is divided left and right and set independently. The face recognition system of claim 1, wherein the left eye of the user's face is located within the left eye position limiting area and the right eye is located within the right eye position limiting area. The method of claim 7, wherein And a size of the left eye position limiting region or the right eye position limiting region can be arbitrarily set and changed. The method of claim 7, wherein And a distance between the left eye position limiting region and the right eye position limiting region can be arbitrarily set and changed. The method according to any one of claims 2 to 4, The predetermined face part is a face outline detected from the user face, And face recognition only when at least a part of the line shapes constituting the face outline are located within a preset face outline restriction area. The method of claim 10, And a size or shape of the face outer limit region may be arbitrarily set and changed. The method of claim 10, The face outer limit region is formed based on a face region minimum limit border and a face area maximum limit border. The size or shape of the face area minimum limit border and the face area maximum limit border may be arbitrarily set and changed. The method according to any one of claims 2 to 4, The predetermined face area is a face outline with respect to both eyes and the user's face, Face recognition is performed only when both eye parts are located within a preset eye position limiting area, and at least a part of a line shape constituting the face outline is located within a preset face outer limiting area. system. The method of claim 13, And a size or a shape of the eye position limiting region or the face outer limiting region can be arbitrarily set and changed. The method according to any one of claims 1 to 4, Further comprising a human body approach detecting unit for detecting the approach of the human body to the front of the camera unit, When the human body approach detection unit detects the approach of the user attempting to recognize the face to maintain a wake-up mode to perform a face recognition, when the user is not detected by maintaining a sleep mode to minimize power consumption Face recognition system. Photographing a user attempting to recognize a face to generate a user image; Detecting both eye parts of the user from the generated user image; Determining whether the detected both eye parts are located within a preset eye position limit area; When both eye parts are located within the eye position limiting area, face recognition is performed by determining whether a user face detected from the user image matches a registered face registered in advance, and the eye parts are located in the eye position. If not located within the restricted area, instructing the user to change the position of the face; Face recognition method comprising a. The method of claim 16, If both eye parts are located within the eye position limiting area, And face recognition when at least a part of the line shapes constituting the face outline detected from the user's face are located within a preset face outline restriction area. Photographing a user attempting to recognize a face to generate a user image; Detecting a shape of a line forming an outline of a user's face in the generated user image; Determining whether at least a part of the detected line shape is located within a preset face outer limit area; If at least a portion of the line shape is located within the face outside restriction area, face recognition is performed by determining whether a user face detected from the user image matches a registered face registered in advance, otherwise face recognition is performed. Inducing the user to change the position of the face without performing; Face recognition method comprising a.

Images