JP2018200592A - Face authentication device, face authentication method, and program - Google Patents

Abstract

To provide a face authentication device, a face authentication method, and a program, capable of obtaining face recognition results with high authentication accuracy.SOLUTION: A face authentication device 1: detects each facial organ (right eyebrow 32a, left eyebrow 32b, right eye 32c, left eye 32d, nose 32e, mouth 32f, jaw 32g, and the like) from a face region 31 contained in an input image 30; generates face organ data 32 defining a position and a shape of each of the detected facial organ with a plurality of feature points; aligns the input face organ data 32 and the facial organ data 82 registered in association with user information in advance with reference to a specific facial organ (for example, jaw 32g); calculates a similarity score; and performs face authentication processing based on similarity. Aligning the input face organ data and the facial organ data with reference to a specific facial organ makes it easier to identify individual differences, thereby obtaining accurate face authentication results.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a face authentication device, a face authentication method, and a program, and more particularly, to a face authentication device, a face authentication method, and a program that use features of facial organs such as eyes, nose, and mouth.

  Conventionally, an authentication system for detecting a facial feature from an image and authenticating a person has been proposed. For example, in the technique of Non-Patent Document 1, feature points of each facial organ (eyebrow, eye, nose, mouth, chin, etc.) included in the face region 301 in the input image 300 are detected and defined by a plurality of feature points. Facial organ data 332 can be generated (FIG. 10A). As described above, using the facial organ data 332 representing the facial features of a person, facial organ data 332 (FIG. 10A) generated from the input image 300 and the facial organ data 382a of a plurality of persons registered. , 382b, 382c,... (FIG. 10B), the degree of similarity of the face can be evaluated. In this way, if facial organ data is used for evaluation of similarity, it is point comparison, so that authentication processing can be performed at a higher speed than when machine learning or the like is used. Further, there is an advantage that learning data collection and learning processing are unnecessary.

  By the way, when evaluating the similarity, it is necessary to align the facial organ data 332 detected from the input image 300 with the registered facial organ data 382a, 382b, 382c,. For example, as disclosed in Non-Patent Document 2, alignment is generally performed using feature points of the entire face. Specifically, as shown in FIG. 11, the parallel movement is performed so that the difference between the corresponding feature points (specifically, the sum of squares of the Euclidean distance between the feature points) is minimized in the entire face. Image processing such as rotation, enlargement / reduction, etc. is performed.

  In addition, the similarity evaluation after alignment is generally performed based on the Euclidean distance between corresponding feature points. For example, an average value of Euclidean distances between the feature points is calculated, and the smaller the average value, the more similar the face is determined.

Vahid Kazemi, 1 other, 2014, "OneMillisecond Face Alignment with an Ensemble of Regression Trees", [online], Royal Institute of Technology, Computer Vision and Active Perception Lab (CVPR), [April 25, 2017 search] , Internet, <URL: http: //www.cv-foundation.org/openaccess/content_cvpr_2014/papers/Kazemi_One_Millisecond_Face_2014_CVPR_paper.pdf> TF Cootes and CJTaylor, 2004, “Statistical Models of Appearance for Computer Vision”, p102 &#8211; 104, [online], [April 25, 2017 search], Internet, <URL: http: // www. face-rec.org/algorithms/AAM/app_models.pdf>

  However, as described above, if alignment is performed so that the difference between corresponding feature points is minimized in the entire face, individual differences are difficult to occur, and the accuracy of authentication is reduced. Furthermore, when only the Euclidean distance between feature points is used as the similarity evaluation index, the similarity evaluation results greatly depend on the organ with many feature points, and even if the same person is used, the similarity evaluation is based on facial expressions. Since the result changes greatly, the accuracy of authentication decreases.

  SUMMARY An advantage of some aspects of the invention is that it provides a face authentication apparatus, a face authentication method, and a program capable of obtaining a face authentication result with high authentication accuracy.

  In order to solve the above-described problem, the first invention associates a plurality of face images with organ data in which the position and shape of each organ of each face image are defined by a plurality of feature points and stores them as registered face data. Storage means; input means for inputting a face image; organ detection means for detecting each organ from a face area included in the face image input to the input means; and organ data for generating organ data of each detected organ Generating means; input organ data that is organ data generated by the organ data generating means; and positioning means for aligning the face image stored in the storage means with reference to a specific organ; A similarity calculation unit that calculates a similarity between the input face image and the face image stored in the storage unit; and an authentication processing unit that performs a face authentication process based on the similarity. A face authentication apparatus according to claim.

  According to the first invention, a plurality of face images and organ data in which the positions and shapes of the organs of each face image are defined by a plurality of feature points are linked and stored in the storage means as registered face data, and input Detecting each organ from the face area included in the face image, generating organ data of each detected organ, aligning the input organ data and the pre-stored face image with reference to a specific organ, The similarity between the registered face image and the stored face image is obtained, and face authentication processing based on the similarity is performed. By performing face image alignment based on a specific organ, individual differences can be easily identified, and an accurate face authentication result can be obtained.

  In the first invention, for example, it is desirable that the similarity calculation means calculates the similarity using feature points for each organ. Since the similarity for each organ is calculated, it becomes easy to identify an individual. Further, the positioning means performs positioning so that a difference between corresponding feature points of the input organ data and the organ data of the registered face data in the specific organ is minimized. Thereby, the alignment of the feature points of the entire face is preferably executed using the feature points of a specific organ. In addition, for example, the positioning unit may perform positioning by performing at least one of image processing among translation, rotation, enlargement, and reduction for one or both of the input organ data and the organ data of the registered face data. Align. Thereby, restrictions such as posture and camera position at the time of photographing can be reduced, and it becomes easy to use.

  Further, the similarity calculation means calculates an average value and variance of Euclidean distances between corresponding feature points in the organ data of the input organ data and the registered face data for each facial organ, and It is desirable to calculate the similarity of the entire face based on the average value and variance of the Euclidean distance. By evaluating not only the Euclidean distance between feature points but also the variance, it becomes easier to identify the same person regardless of changes in facial expressions, and the authentication accuracy is improved.

  The number of feature points of the organ data is preferably defined for each organ. As a result, features can be extracted with the optimum number of feature points according to each organ, and facial features can be easily expressed. As a result, it is possible to obtain a robust face authentication result that makes it easy to identify individual differences and is less dependent on facial expressions.

  According to a second aspect of the invention, the computer associates a plurality of face images with organ data in which the positions and shapes of the organs of the face images are defined by a plurality of feature points and stores them in the storage unit as registered face data. A step of inputting a face image, a step of detecting each organ from a face region included in the input face image, a step of generating organ data of each detected organ, and an input organ which is the generated organ data A step of aligning the data and the face image stored in the storage unit with reference to a specific organ, and the similarity between the registered face image stored in the storage unit and the face image stored in the storage unit A face authentication method comprising: a step of calculating; and a step of performing face authentication processing based on the similarity.

  According to the second invention, a plurality of face images and organ data in which the positions and shapes of the respective organs of the face images are defined by a plurality of feature points are linked and stored in the storage unit as registered face data, and input Detecting each organ from the face area included in the face image, generating organ data of each detected organ, aligning the input organ data and the pre-stored face image with reference to a specific organ, The similarity between the registered face image and the stored face image is obtained, and face authentication processing based on the similarity is performed. By performing face image alignment based on a specific organ, individual differences can be easily identified, and an accurate face authentication result can be obtained.

  A third invention relates to a storage means for storing a computer as a registered face data by associating a plurality of face images with organ data in which the positions and shapes of the organs of the face images are defined by a plurality of feature points. Input means for inputting an image, organ detection means for detecting each organ from a face area included in the face image input to the input means, organ data generating means for generating organ data of each detected organ, and organ data generation Positioning means for aligning input organ data, which is organ data generated by the means, and the face image stored in the storage means with reference to a specific organ, the input face image that has been aligned, and the storage means A program for functioning as similarity calculation means for calculating the similarity to the face image stored in the image, and authentication processing means for performing face authentication processing based on the similarity That.

  According to the third invention, it is possible to cause a computer to function as the face authentication device of the first invention.

  According to the present invention, it is possible to provide a face authentication device, a face authentication method, and a program capable of obtaining a face authentication result with high authentication accuracy.

The block diagram which shows the function structure of the face authentication apparatus 1 which concerns on this invention. Internal configuration diagram of computer functioning as face authentication device 1 The figure which shows the data structural example of the registration facial organ data 8 The flowchart which shows the flow of the authentication process which the face authentication apparatus 1 performs The figure explaining the process which detects the facial organ data 32 from the image 30 The figure explaining alignment with the input facial organ data 32 and the registration facial organ data 82 Flow chart showing the flow of similarity evaluation processing The figure explaining similarity calculation for every facial organ The figure explaining the comparison of the corresponding feature points of the input facial organ data 32d and the registered facial organ data 82d The figure explaining the process which detects the facial organ data 332 from the input image 300, and compares with the registered facial organ data 382a, 382b, ... The figure explaining the alignment of the conventional facial organ data

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

  FIG. 1 is a block diagram showing a functional configuration of a face authentication device 1 according to the present invention, and FIG. 2 is a diagram showing an example of an internal configuration of a computer that functions as the face authentication device 1.

  As shown in FIG. 1, the face authentication device 1 includes an input unit 2, a face region detection unit 3, a face organ detection unit 4, a registration unit 5, a similarity evaluation unit 6, an output unit 7, a registration processing unit 9, and A storage unit 12 is provided. The storage unit 12 stores registered face data 8.

  The input unit 2 inputs the image 30 to the face authentication device 1. An input image 30 (hereinafter referred to as an input image) is an image including a face area (face image), and may be a still image or a moving image. The input unit 2 may acquire the image 30 captured by the camera 20 directly from the camera 20 or may acquire a face image stored in the storage unit 12 or a recording medium.

  The face area detection unit 3 determines whether or not a human face is included in the input image 30 input from the input unit 2, and detects a face area 31 from the input image 30 when a human face is detected ( (See FIG. 5).

The face organ detection unit 4 detects each organ from the face region 31 detected by the face region detection unit 3. The face organ detection unit 4 detects predetermined feature points from the image information of the detected face region 31 and estimates the position and shape of each face organ. The facial organs are organs included in the face, such as the right eyebrow 32a, the left eyebrow 32b, the right eye 32c, the left eye 32d, the nose 32e, the mouth 32f, and the jaw 32g. A known technique may be used to detect the facial organs. For example, the method of Non-Patent Document 1 is known. (Non-Patent Document 1: Vahid Kazemi and 1 other, "One Millisecond Face Alignment with an Ensemble of Regression
Trees ", [online], Royal Institute of Technology, Computer Vision and Active Perception Lab (CVPR), <URL: http://www.cv-foundation.org/openaccess/content_cvpr_2014/papers/Kazemi_One_Millisecond_Face_2014_CVPR_paper.pdf>). According to the technique of Non-Patent Document 1, organ data 32 (hereinafter referred to as face organ data) in which the positions and shapes of the facial organs 32a to 32g are defined by a plurality of feature points can be generated from the face region 31. In the present invention, the facial organ detection unit 4 only needs to detect the facial organs 32a to 32g from the image 30 and generate facial organ data 32 in which the positions and shapes of the facial organs 32a to 32g are defined by a plurality of feature points. The detection method of the facial organs 32a to 32g is not limited to the method of Non-Patent Document 1. The facial organ data 32 will be described later.

  The facial organ detection unit 4 outputs the facial organ data generated from the input image 30 to the registration processing unit 9 in the registration mode and to the registration unit 5 in the face authentication mode. The registration mode is a processing mode for registering the face image and face organ data of the user who uses the face authentication device 1 in association with each other. The face authentication mode is a processing mode for executing the face authentication process of the user of the face authentication device 1. Whether the face authentication apparatus 1 executes the registration mode or the face authentication mode is assumed to be selected in advance by the operator.

  The registration processing unit 9 associates the facial organ data generated by the facial organ detection unit 4 in the registration mode with the original facial image 81 as the facial organ data 82 and stores it in the storage unit 12 (see FIG. 3). Further, user information (hereinafter referred to as user information 83) may be stored in the storage unit 12 in association with the face image 81 and the facial organ data 82. The user information 83 is information input to the face authentication apparatus 1 by the user in the registration process, and includes, for example, attribute information such as the user's name, age, and sex.

  FIG. 3 is a diagram illustrating an example of registered face data 8 stored in the storage unit 12. As shown in FIG. 3, the registered face data 8 includes a user ID 80 that is user identification information, a user face image 81, face organ data 82 generated from the face image 81, and user attributes. Information 83 is linked and stored. User attribute information includes name, age, gender, and the like. In the storage unit 12, a plurality of user face images 81, face organ data 82, and user information 83 are linked and stored as registered face data 8.

  On the other hand, in the face authentication mode, the facial organ data detected by the facial organ detection unit 4 is input to the alignment unit 5. The facial organ data detected by the facial organ detection unit 4 is referred to as input facial organ data 32.

  The alignment unit 5 identifies an input image 30 (referred to as an input face image 30) input by the input unit 2 in the face authentication mode and a face image (registered face image 81) stored in advance in the storage unit 12. Align with the facial organs as a reference. The alignment is performed by (1) aligning the generated face organ data (input face organ data 32) detected from the input face image and the face image (registered face image 81) stored in advance in the storage unit 12. (2) The face images (the input face image 30 and the registered face image 81) may be aligned, or (3) the facial organs stored in advance in the input face image 30 and the storage unit 12 Data (registered facial organ data 82) may be aligned. In any case, in the present invention, the registration unit 5 uses the input face image 30 (or input face organ data 32) and the registered face image 81 (or registered face organ data 82) as a specific face organ (for example, jaw). ) For reference. As a specific example, the registration unit 5 determines the difference between the corresponding feature points in the specific facial organs of the input facial organ data 32 and the registered facial organ data 82 (specifically, the Euclidean distance between the feature points). Alignment is performed so that the sum of squares) is minimized. The alignment of the facial organ data 32 and 82 will be described later.

  The similarity evaluation unit 6 calculates the similarity between the input face image 30 registered by the registration unit 5 and the registered face image 81 in the face authentication mode. The similarity evaluation unit 6 calculates the similarity between the input face image 30 and the registered face image 81 using the feature points for each facial organ. Specifically, the average value and variance of the Euclidean distance between corresponding feature points in the input facial organ data 32 and the registered facial organ data 82 are calculated for each facial organ. Further, the similarity score of the entire face is calculated based on the average value and variance of the calculated Euclidean distances of all facial organs. A specific calculation method will be described later.

  The similarity evaluation unit 6 compares the calculated similarity score with a predetermined threshold, and if the similarity score is equal to or lower than the predetermined threshold, the input facial organ data 32 and the registered facial organ data 82 are similar. It is determined that In the face authentication mode, alignment and similarity evaluation of all registered facial organ data 82 and input facial organ data 32 registered in the storage unit 12 are performed. In the similarity evaluation, a person associated with registered facial organ data having a similarity score that is equal to or lower than a predetermined threshold and having the smallest similarity score is authenticated as the person. If the similarity score is larger than a predetermined threshold value, authentication fails. A specific example of similarity calculation and authentication processing in the similarity evaluation unit 6 will be described later.

  The output unit 7 outputs the evaluation result (authentication processing result) in the similarity evaluation unit 6. For example, the output is a notification of the authentication result to the login processing unit that executes the login process if it is an authentication process at the time of login of the PC. This is a notification of the authentication result to the unlocking processing unit to be executed. Note that the output method is not limited to these, and the evaluation result of the similarity evaluation unit 6 may be notified by any method.

  FIG. 2 is a diagram illustrating a configuration example of a computer that functions as the face authentication apparatus 1. As shown in FIG. 2, the face authentication device 1 includes a control unit 11, a storage unit 12, a media input / output unit 13, a peripheral device I / F unit 14, an input unit 15, a display unit 16, a communication control unit 17, and the like. 18 is connected. A camera 20 is connected to the peripheral device I / F unit 14. When the computer functions as the face authentication apparatus 1, the computer control unit 11 is realized by executing a program describing the processing shown in FIG.

  The control unit 11 includes a CPU (Central Processing Unit), a ROM (Read Only Memory), a RAM (Random Access Memory), and the like. The CPU calls a program stored in the storage unit 12, ROM, recording medium or the like to a work memory area on the RAM and executes it, and drives and controls each unit connected via the bus 18. The ROM permanently holds a computer boot program, a program such as BIOS, data, and the like. The RAM temporarily holds the loaded program and data, and includes a work area used by the control unit 11 to perform various processes. The control unit 11 functions as each unit of the face authentication device 1 by reading and executing the program.

  The storage unit 12 is a storage device such as a hard disk drive, for example. The storage unit 12 stores a program executed by the control unit 11, data necessary for program execution, an operating system, and the like. These program codes are read by the control unit 11 as necessary, transferred to the RAM, and read and executed by the CPU. The storage unit 12 stores registered face data 8 in which registered face images 81 that are face images of each user are associated with registered face organ data 82 that is face organ data generated from the face images. (See FIG. 3).

  The media input / output unit 13 is, for example, a drive device for various recording media (media) such as a CD, a DVD, and an MO, and inputs / outputs (writes / reads) data to / from the media.

  The peripheral device I / F (interface) unit 14 is a port for connecting a peripheral device, and transmits / receives data to / from the peripheral device via the peripheral device I / F unit 14. The peripheral device I / F unit 14 is configured by a USB or the like, and usually includes a plurality of peripheral devices I / F. The connection form with the peripheral device may be wired or wireless.

The input unit 15 is an input device such as a touch device, a pointing device such as a keyboard and a mouse, and a numeric keypad, and outputs input data to the control unit 11.
The display unit 16 includes a display device such as a liquid crystal panel or a CRT monitor, and a logic circuit (video adapter or the like) for executing display processing in cooperation with the display device, and is input under the control of the control unit 11. Display information is displayed on a display device. The input unit 15 and the display unit 16 may be a touch panel display (operation unit 25) in which an input device such as a touch panel is integrally provided on the display screen.

The communication control unit 17 includes a communication control device, a communication port, and the like, and controls communication with the network 10 and the like. The network 10 includes a LAN (Local Area Network), a WAN (Wide Area Network) connected to a wider area, or a public communication line such as the Internet, a base station, and the like. The network 10 may be wired or wireless. The face authentication device 1 can access a server (not shown) via the network 10 and can transmit and receive various programs and data.
The bus 18 is a path that mediates transmission / reception of control signals, data signals, and the like between the devices.

  Next, a flow of face authentication processing executed by the face authentication device 1 will be described with reference to FIG. FIG. 4 is a flowchart for explaining the flow of processing in the face authentication mode. The control unit 11 of the face authentication device 1 reads the program and data related to the process shown in FIG. 4 from the storage unit 12, and executes the process based on the program and data. It is assumed that registered face data 8 of a plurality of users are registered in the storage unit 12 before the face authentication process is executed. The facial organ data 82 registered in the registered face data 8 may be created by the same device as the face authentication device 1 that executes the following face authentication processing, or created by another device and stored in the storage unit 12. It may be stored in

  The control unit 11 (input unit 2) of the face authentication apparatus 1 acquires an image captured by the camera 20 or an image stored in the storage unit 12 as the input image 30 (step S101). The input image 30 acquired in step S101 may be a still image or a moving image. In the case of a moving image, the following processing is performed using the image of each frame included in the moving image as the input image 30.

  The control unit 11 (face area detection unit 3) detects a human face area 31 from the input image 30 (step S102). The control unit 11 (face organ detection unit 4) detects the face organs 32a to 32g from the detected face region 31 and generates face organ data 32 (step S103). The facial organ data 32 generated in step S103 is referred to as input facial organ data 32.

  FIG. 5 is a diagram for explaining a process of generating facial organ data 32 from the input image 30. As described above, the control unit 11 recognizes the face area 31 from the image 30 (an image of each frame in the case of acquiring a video) taken by the camera 20, detects the face organs 32a to 32g from the face area 31, Facial organ data 32 is generated. As a method for generating the facial organ data 32 from the image 30, for example, a known algorithm shown in Non-Patent Document 1 described above may be used. The method for generating the facial organ data 32 is not limited to the method described in Non-Patent Document 1. The present invention may generate the facial organ data 32 using other methods.

  FIG. 5A is a diagram showing a face area 31 and face organ data 32 detected from the input image 30. As shown in FIG. 5B, the facial organ data 32 has a plurality of positions and shapes of each facial organ (right eyebrow 32a, left eyebrow 32b, right eye 32c, left eye 32d, nose 32e, mouth 32f, jaw 32g). Data defined by feature points. The number of feature points of each facial organ 32a to 32g is defined for each facial organ. Specifically, for example, the right eyebrow 32a, the left eyebrow 32b, the right eye 32c, and the left eye 32d are 20 feature points. The nose 32e is defined by 17 feature points. The mouth 32f is defined by 54 feature points. The jaw is defined by 41 feature points. In addition, the number of these feature points is an example, and other numbers may be used.

  In step S103, if the face cannot be caught in front due to the installation position of the camera 20, the movement of the person's face, etc., the image (frame) may be excluded from the face organ detection target. .

  Next, the control unit 11 acquires, from the storage unit 12, registered facial organ data 82, which is facial organ data of the registered facial image 81 to be compared with the input facial organ data 32 generated from the input image 30 (step S104). It is assumed that the registered facial organ data 82 of the user is stored in the storage unit 12 as shown in FIG.

  The control unit 11 (alignment unit 5) aligns the facial organ data 32 (input facial organ data 32) generated in step S103 and the registered facial organ data 82 called from the storage unit 12 in step S104 (step S104). S105).

  In the alignment process in step S105, the control unit 11 aligns the input facial organ data 32 and the registered facial organ data 82 with reference to a specific facial organ. That is, alignment is performed so that a difference between corresponding feature points in a specific face organ in the input face organ data 32 and the registered face organ data 82 (a sum of squares of the Euclidean distance between the feature points) is minimized. .

  The specific facial organ used as a reference for alignment is preferably an organ that does not easily change depending on facial expressions, such as the jaw shown in FIG. Further, which facial organ is the specific facial organ may be changed by a setting operation performed in advance by the operator of the face authentication apparatus 1. In this way, by comparing the input facial organ data 32 and the registered facial image 81 (registered facial organ data 82) with a specific one organ (for example, the jaw) as a reference, the individual differences of the faces are compared. It becomes easy to do.

  In the alignment process of step S105, the control unit 11 performs at least one or more of translation, rotation, enlargement, and reduction on one or both of the input facial organ data 32 and the registered facial organ data 82. Image processing is performed, and alignment is performed so that a difference between corresponding feature points of a specific facial organ (sum of squares of Euclidean distance between feature points) is minimized. In this way, the alignment of the feature points of the entire face is preferably executed using the feature points of a specific facial organ. In addition, since alignment is performed by performing at least one of image processing among translation, rotation, enlargement, and reduction, restrictions on posture and camera position at the time of shooting can be reduced, and it is easy to use. It will be a thing.

  FIG. 6 shows a state in which the input facial organ data 32 and the registered facial organ data 82 are aligned based on the jaw 32g.

  In the registration process in step S105, an example in which the input facial organ data 32 and the registered facial organ data 82 are registered has been shown. However, the present invention is not limited to this example, and the input facial organ data 32, the registered facial image 81, and the like. May be registered, the face images (input face image 30 and registered face image 81) may be aligned, or the input face image 30 and registered face organ data 82 may be aligned. . In any case, the input face image 30 (or input face organ data 32) and the registered face image 81 (or registered face organ data 82) are aligned based on a specific face organ (for example, chin).

  Next, the control unit 11 (similarity evaluation unit 6) evaluates the similarity between the registered input face image 30 and the registered face image 81 (step S106). In the similarity evaluation process, first, the similarity between the registered input facial organ data 32 and the registered facial organ data 82 is calculated. When calculating the similarity, the control unit 11 calculates the average value and variance of the Euclidean distance between corresponding feature points in the input facial organ data 32 and the registered facial organ data 82 for each facial organ. Next, the similarity of the entire face is comprehensively evaluated using the weighted sum of the scores of similarity for each facial organ.

  Since the shape of each facial organ is almost constant for the same person, there is no variation in the value of the distance between corresponding feature points, and the variance is predicted to be small. In this way, the average value and variance of the distance between feature points for each facial organ are calculated, and if the overall similarity is comprehensively evaluated, the similarity of each facial organ alone is evaluated and the entire face is evaluated. The degree of similarity can also be evaluated. Thereby, the same person can be identified without depending on the change in facial expression. Further, the similarity can be evaluated for the entire face without being biased toward a specific facial organ.

The similarity evaluation process in step S106 will be specifically described with reference to FIG.
As shown in FIG. 7, first, K, which is a value indicating the facial organ to be calculated, is initialized (K = 1; step S201), and the input facial organ data 32 and the registered facial organ data 82 for the facial organ K are obtained. calculating a Euclidean distance _{d n} between the corresponding feature points n (step S202).

  The position coordinates of the feature point n in the input facial organ data 32 and the registered facial organ data 82 are respectively

When the Euclidean distance d _n between the images of the feature points n (between input face organ data 32 and the registered face organ data 82) can be expressed as the following equation (1).

Here, if the set K is placed for the facial organ K as K = {k | k is a feature point included in the facial organ K}, Euclidean between images at the characteristic point k included in the facial organ K. The average μ _K of distance is as shown in the following formula (2).

Here, as shown in FIG. 8, the facial organ K is the number of the facial organ to be calculated. For example, the right eyebrow K = 1, the left eyebrow K = 2, the right eye K = 3, and the left eye K = 4, the nose K = 5, the mouth K = 6, the chin K = 7, and so on. num _K is the number of feature points included in the facial organ K.

Next, the control unit 11 obtains a variance sigma _K ² of the Euclidean distance between the corresponding feature points of the face organs K in the input face organ data 32 registered face organ data 82 (step S203). Variance sigma _K ² of the Euclidean distance between the corresponding feature points of the face organs K in the input face organ data 32 registered face organ data 82 is given by the following equation (3). The variance σ _K ² is a value that decreases as the facial organ shape is closer.

  FIG. 9 is a diagram showing feature points of the right eyes 32d and 82d in the input facial organ data 32 and the registered facial organ data 82. As shown in FIG. When the input facial organ data 32 and the registered facial organ data 82 belong to the same person, as shown in FIG. 9, the shape of the facial organs are almost the same, but the positions may be shifted due to changes in facial expressions. is there. In such a case, it can be assumed that the value of the distance between the corresponding feature points of the input facial organ data 32 and the registered facial organ data 82 has a small variation (variance). By using variance for evaluation of similarity, individual differences of faces can be more accurately evaluated.

  The control unit 11 repeats the calculation of the Euclidean distance and the variance between the corresponding feature points for all the facial organs K (Step S204; No → Step S205 → Step S202 to Step S203).

  When the Euclidean distance and variance between corresponding feature points are calculated for all facial organs K (step S204; Yes), the control unit 11 calculates the overall similarity score of the entire face (step S206). The similarity score (score) is, for example, the following formula (4).

In Equation (4), μ _K and σ _K ² are the average value and variance of Euclidean distances between images at feature points included in the facial organ K. w _K1 and w _K2 are weight parameters. The value of the weight parameter is an appropriate value determined experimentally. Face is a set of detected facial organs. For example, in the example shown in FIG. 7, Face = {right eyebrow, left eyebrow, right eye, left eye, nose, mouth, chin}.

  The similarity score calculated by the above equation (4) becomes smaller as the input facial organ data 32 and the registered facial organ data 82 are more similar.

  The control unit 11 outputs the similarity score value calculated in step S206 as a similarity evaluation result between the input face image 30 and the registered face image 81 acquired in step S104. The control unit 11 repeatedly executes the processing of Step S104 to Step S106 (acquisition of registered face image 81 (registered face organ data 82), alignment processing, similarity evaluation processing) for all registered face images 81 registered. To do. Thereby, the control unit 11 obtains a similarity score between the input face image 30 and the registered face image 81.

  In the authentication process of step S107 in FIG. 4, the control unit 11 performs personal authentication based on the similarity evaluation result (similarity score) between the input face image 30 and the registered face image 81 (step S107). In the personal authentication, the control unit 11 determines which registered face image 81 the person of the input image 30 is associated with. The control unit 11 authenticates a person who is associated with the registered face image 81 having a similarity score equal to or lower than a predetermined threshold and having the smallest similarity score as the person. If the similarity score is greater than a predetermined threshold, it is determined that face authentication has failed.

  The control part 11 (output part 7) outputs the authentication result of step S107 (step S108 of FIG. 4). For example, the output is a notification of the authentication result to the login processing unit if it is an authentication process at the time of login of the PC, and a notification of the authentication result to the unlocking processing unit if it is a face authentication process in gate unlocking etc. It is. Note that the result output method is not limited to these, and any output method may be used.

  As described above, the face authentication device 1 according to the present embodiment is configured so that each facial organ (right eyebrow 32a, left eyebrow 32b, right eye 32c, left eye 32d, nose 32e, Mouth 32f, jaw 32g, and the like), and facial organ data 32 in which the positions and shapes of the detected facial organs 32a to 32g are defined by a plurality of feature points is generated. The registered facial organ data 82 of the registered facial image 81 registered in association with the information 83 is registered with reference to a specific facial organ (for example, the jaw 32g), a similarity score is obtained, and based on the similarity score Perform face authentication processing. By performing alignment based on a specific facial organ, individual differences can be easily identified, and an accurate face authentication result can be obtained. The similarity score is a value obtained by integrating the average value and variance of Euclidean distances between images at feature points included in each facial organ K in all facial organs. By evaluating not only the Euclidean distance between corresponding feature points but also the variance, the same person can be easily identified regardless of the change in facial expression, and the authentication accuracy is improved.

  The preferred embodiments of the face authentication apparatus and the like according to the present invention have been described above with reference to the accompanying drawings, but the present invention is not limited to such examples. For example, in the above-described embodiment, the example in which the computer executes the face authentication process has been described. However, the present invention is not limited to this, and can be applied to, for example, a process in which a robot specifies a person from a face. It is. In addition, it is obvious for those skilled in the art that various changes or modifications can be conceived within the scope of the technical idea disclosed in the present application, and these naturally belong to the technical scope of the present invention. It is understood.

1. Face recognition device (computer)
2 ……………… Input section 3 ……………… Face area detection section 4 ……………… Face organ detection section 5 ……………… Alignment section 6 ……………… Similarity Evaluation unit 7 ……………… Output unit 8 ……………… Registered face data 9 ……………… Registration processing unit 10 ……………… Network 11 …………… Control unit 12 ……… …… Storage unit 13 ……………… Media input / output unit 14 …………… Peripheral device I / F unit 15 …………… Input unit 16 …………… Display unit 17 …………… Communication control Part 18 …………… Bus 30 …………… Image 31 …………… Face area 32 …………… Facial organ data (input facial organ data)
32a to 32g ......... Facial organs 80 ... …… User ID
81 ……………… Face image (Registered face image)
82 .......... Facial organ data (Registered facial organ data)
83 ………… User attribute information

Claims (8)

Storage means for storing a plurality of face images and organ data in which the position and shape of each organ of each face image are defined by a plurality of feature points and storing them as registered face data;
An input means for inputting a face image;
Organ detection means for detecting each organ from a face area included in the face image input to the input means;
Organ data generating means for generating organ data of each detected organ;
Alignment means for aligning input organ data, which is organ data generated by the organ data generation means, and a facial image stored in the storage means with reference to a specific organ;
Similarity calculation means for calculating the similarity between the registered face image and the face image stored in the storage means;
Authentication processing means for performing face authentication processing based on the similarity;
A face authentication apparatus comprising:   The face authentication apparatus according to claim 1, wherein the similarity calculation unit calculates the similarity using a feature point for each organ.   The positioning means performs positioning so that a difference between corresponding feature points between the input organ data in the specific organ and organ data of registered face data stored in the storage means is minimized. The face authentication apparatus according to claim 1, wherein:   The alignment unit performs image processing of at least one of translation, rotation, enlargement, and reduction on one or both of the input organ data and the organ data of the registered face data. The face authentication apparatus according to claim 1, wherein alignment is performed.   The similarity calculation means calculates, for each organ, an average value and variance of Euclidean distances between corresponding feature points in the organ data of the input organ data and the registered face data, and calculates the Euclidean distance of all organs. The face authentication apparatus according to claim 1, wherein the degree of similarity of the whole face is calculated based on the average value and the variance.   6. The face authentication apparatus according to claim 1, wherein the number of feature points of the organ data is defined for each organ. Computer
Linking a plurality of face images and organ data defined by a plurality of feature points to the position and shape of each organ of each face image and storing them as registered face data in a storage unit;
A step of inputting a face image;
Detecting each organ from the face area included in the input face image;
Generating organ data for each detected organ;
Aligning the input organ data, which is the generated organ data, with the face image stored in the storage unit with reference to a specific organ;
Calculating a similarity between the input face image that has been aligned and the face image stored in the storage unit;
Performing face authentication processing based on the similarity;
A face authentication method comprising: Computer
Storage means for storing a plurality of face images and organ data in which the positions and shapes of the respective organs of each face image are defined by a plurality of feature points and storing them as registered face data;
An input means for inputting a face image;
Organ detection means for detecting each organ from a face region included in the face image input to the input means;
Organ data generating means for generating organ data of each detected organ;
Alignment means for aligning input organ data, which is organ data generated by the organ data generation means, and a facial image stored in the storage means with reference to a specific organ;
Similarity calculation means for calculating the similarity between the input face image that has been aligned and the face image stored in the storage means;
Authentication processing means for performing face authentication processing based on the similarity,
Program to function as.

Images