JP2014116033A - Personal authentication device, personal authentication method, camera, program, and storage medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize a personal authentication technique which can accurately perform personal authentication at high speed even when a facial expression of an authentication target person has changed.SOLUTION: A personal authentication device determines a facial expression by a face area of a person included in input image data, selects feature data corresponding to the determined facial expression from a plurality of pieces of feature data for personal authentication that are stored in storage means, and compares feature data of the face area by which the facial expression is determined with the selected feature data in order to authenticate the person. Further, the personal authentication device determines the face area of the same person by comparing information on the face area among a plurality of frames, and when feature data corresponding to a new facial expression which is not yet stored is detected from the face area of the person who has already been authenticated with the image data of a previous frame, stores the feature data as the feature data corresponding to the new facial expression of the person in storage means.

Description

  The present invention relates to a technique for performing personal authentication by detecting a human face area from image data.

  Conventionally, a technique for performing personal authentication using a face detection function is known. This technology extracts feature data of different face areas for each individual from detected face image data and compares it with feature data registered in advance to identify whether the detected face is a registered person or not It is. However, since the feature data is affected by the facial expression, the accuracy of personal authentication may be deteriorated when the facial expression is different from that at the time of registration. In order to avoid this, there is a known method of detecting facial expression in the face area, determining whether the facial expression is valid for personal authentication from the detected facial expression, and if invalid, performing personal authentication with new image data. (For example, refer to Patent Document 1).

Japanese Patent Laid-Open No. 06-119433

  However, in the method described in Patent Document 1, personal authentication cannot be performed until the facial expression becomes valid. Therefore, it takes time until the authentication, and if an effective facial expression does not appear, authentication is not performed. It was impossible.

  The present invention has been made in view of the above problems, and realizes a personal authentication technique capable of performing personal authentication at high speed and with high accuracy even when a facial expression changes.

  In order to solve the above-described problem, an authentication apparatus according to the present invention includes facial expression determination means for determining facial expression from a human face area included in input multiple frames of image data, and multiple expressions included in the multiple frames of image data. A means for comparing the information of the face areas and determining a face area of the same person among the plurality of face areas, and a storage means for storing a plurality of facial expressions and feature data for authenticating the individual in association with each other The selection means for selecting feature data corresponding to the facial expression determined by the facial expression determination means, the feature data of the face area for which the facial expression is determined by the facial expression determination means, and the feature data selected by the selection means By comparing, the authentication means for authenticating the individual and the individual in the image data of the previous frame in the face area included in the image data of the new frame are authenticated. When feature data corresponding to another facial expression of the individual that is not stored in the storage means is detected from a face region determined to be the same person as the region, the feature data corresponds to another facial expression of the individual. Registering means for storing the characteristic data in the storage means.

  According to the present invention, even when the facial expression changes, the personal authentication is performed by selecting appropriate feature data corresponding to the facial expression, so that the authentication accuracy can be improved.

The block diagram which shows the structure of the personal authentication apparatus of 1st Embodiment which concerns on this invention. FIG. 3 is a diagram illustrating a personal database according to the first embodiment. The flowchart which shows the personal authentication process by 1st Embodiment. The block diagram which shows the structure of the personal authentication apparatus of 2nd Embodiment. The flowchart which shows the personal authentication process by 2nd Embodiment. The figure which illustrates a feature point. The figure which illustrates the feature point in the case of a smile.

  EMBODIMENT OF THE INVENTION Below, the form for implementing this invention with reference to an accompanying drawing is demonstrated in detail. The embodiment described below is an example for realizing the present invention, and should be appropriately modified or changed according to the configuration of the apparatus to which the present invention is applied and various conditions. The present invention is described below. It is not limited to the form.

  [First Embodiment] FIG. 1 is a block diagram showing a schematic configuration of a personal authentication apparatus according to an embodiment of the present invention. The personal authentication device according to the present embodiment inputs image data from the image input unit 1, and the face detection unit 2 detects the position and size of a person's face area from this image data. A known method may be used for face detection. For example, the face detection unit 2 extracts the shape corresponding to the constituent elements of the face area such as the nose, mouth, and eyes from the input image data, and the nose and the mouth exist on the extension line that passes between the eyes. Detect areas. Then, the size of the face is estimated from the size of both eyes and the distance between them, and the region surrounded by the estimated size region is defined as the face region with reference to the position corresponding to the center of the nose.

  The face detection unit 2 extracts feature data from the detected face area. For example, as disclosed in Japanese Patent Application Laid-Open No. 2005-266981, the feature data is information regarding specific shapes of facial components such as mouth, eyes, eyebrows, and nose, and positions of these components. including. Here, the feature data can be extracted from the input image data of the face area by using, for example, a technique such as edge detection using a neural network or a spatial filter. Of course, not only information on the shape and position but also information on saturation and hue may be included in the feature data. The more feature data in one face, the more detailed the state of the face can be analyzed, and the accuracy of facial expression determination and personal authentication using this feature data is improved.

  In the image data including a plurality of frames such as a moving image, the same person determination unit 3 detects a face every frame of a certain period, and determines which face and which face are the same person. To do. For example, when a plurality of faces are detected from image data of a certain frame and one or more faces are detected from image data of another frame, the same person determination unit 3 determines the size and position of each face. Consider similar things to be the same person.

  The facial expression determination unit 4 determines what facial expression the face has based on the feature data extracted by the face detection unit 2. Examples of the types of facial expressions to be determined include no facial expressions, smiles, and eyes closed. Based on the determination result, the database selection unit 5 selects and reads out feature data corresponding to the facial expression from the feature data for personal authentication registered in the personal database unit 8 in advance. The authentication unit 6 performs personal authentication by comparing / collating the feature data selected by the database selection unit 5 with the feature data of the face area extracted by the face detection unit 2, and the result output unit 7 performs the authentication result. Is output.

  This personal authentication device may be constituted by a single device or a system composed of a plurality of devices. For example, all the configurations from the image input unit 1 to the personal database unit 8 may be provided inside a single digital camera or digital video camera. Alternatively, only the image input unit 1 may be provided inside the digital camera or digital video camera, and the other may be provided in an external computer that can communicate with the digital camera or digital video camera. Furthermore, all of the personal database unit 8 from the image input unit 1 is shared by a plurality of computers on the network so that the computer having the image input unit 1 receives image data from another external device or recording medium. You may comprise.

  In the present embodiment, as feature data for personal authentication, coordinates of 23 feature points are used as shown in FIG. In order to actually perform personal authentication, more feature points are required, but here, in order to simplify the description, the description will be made assuming that 23 feature points are used. If the face is tilted in the image data, these 23 feature points are obtained after rotating the image data so as to eliminate this tilt. The coordinates of these feature points are calculated by normalizing the positions of the eyes, nose, mouth, eyebrows and the like extracted from the image data by the face detection unit 2, for example, by the distance between the eyes. The This coordinate is based on the position of the end of the nose. The authentication unit 6 sets the coordinates of each feature point calculated from the input image data as Pi (i = 1, 2,..., 23), and the feature points of the person registered in the personal database unit 8 in advance. The absolute value sum S = Σ | Pi−P′i | of the difference from the coordinate P′i is obtained. It is considered that the smaller the absolute value sum S, the higher the possibility that the person who is the detection target and the person registered in advance are the same person. The authentication unit 6 determines that the absolute value sum S of a person determined to be most likely is equal to or less than a preset threshold value, and determines that there is no corresponding person if the value is greater than the threshold value. To do.

  The method for obtaining the absolute value sum S is one example of a method for performing personal authentication, and the personal authentication may be performed by another method. For example, an individual may be identified from the pattern of changes in the position and shape of eyes and mouth when changing from expressionless to smiling, and a large number of image data with different resolutions are generated from the face image data. Separately, personal authentication may be performed, and a final personal authentication result may be obtained from the comprehensive result. That is, the same effect as that of the present embodiment can be obtained even if other methods are used as long as the configuration is such that it is verified with the data registered in the personal database unit 8 in advance and determined to be the most likely person. Can be obtained.

  Here, the facial expression determination by the facial expression determination unit 4 can be realized by detecting temporal changes in the relative positions of feature points in the same face. For example, in the case of a smile, as compared with the expressionless face shown in FIG. 6, the position of the feature point changes, such as narrowing the eyes and raising the corner of the mouth as shown in FIG. In this embodiment, the distance between the feature points 9 and 11 at the center upper and lower ends of the left eye and the distance between the feature points 13 and 15 at the center upper and lower ends of the right eye toward the distance when there is no expression. If the positions of the feature points 6 and 7 at the left and right ends of the mouth with respect to the feature points 4 and 5 at the center upper and lower ends of the mouth are raised by a predetermined ratio or more compared to the expressionless state, the face is determined to be a smile. Furthermore, the positions of the feature points 4 to 7 of the mouth do not satisfy the smile condition, and the distance between the feature points 9 and 11 of the eyes and the distance between the feature points 13 and 15 with respect to the distance when there is no expression If it becomes smaller, it is determined that the eyes are closed.

  The facial area is determined when there is no expression. For example, when the amount of change in the shape of the mouth does not exceed a predetermined threshold for a predetermined period, it is determined that the facial area at that time is in an expressionless state. Just do it. Alternatively, a value obtained by averaging feature points of several consecutive frames for the same face region may be regarded as an expressionless state. Alternatively, the feature points obtained between several consecutive frames for the same face area are compared, and the relative value of them automatically determines which face in which frame is expressionless and in which frame the face is smiling. You may make it discriminate | determine. Furthermore, conditions are set for the degree of opening of the eyes and the shape of the mouth for each facial expression, which condition is determined for the face of each input frame image, and the facial expression is determined independently for each frame. It may be. If it is the structure which determines an expression independently for every flame | frame, an expression can be determined also from the still image which consists of a single frame. In this way, facial expression determination is different from personal authentication that compares with feature points registered in advance for each person, and the shape of the parts such as eyes and mouth that make up the face extracted from the image data depends on specific conditions. This is done by determining whether or not

  Here, as described above, in the personal authentication, the feature point detected from the image data is compared with the feature point registered in advance in the personal database unit 8 for each person, and the absolute value sum S is obtained. . However, when the facial expression of the person changes, the coordinates of the feature points detected from the image data change, and the value of the absolute value sum S obtained during personal authentication varies greatly depending on the facial expression. There is a problem that accuracy is lowered. On the other hand, in the personal authentication device of the present embodiment, accuracy can be improved by performing personal authentication as follows.

  Characteristic data for personal authentication registered in the personal database unit 8 in advance is registered for each person and facial expression to be personally authenticated as shown in FIG. It is only necessary to specify facial expressions at the time of registration, cause the subject to make facial expressions, and register the feature data extracted for each photographed facial expression. Alternatively, the facial expression may be automatically determined by the method described above from the image data in which the subject arbitrarily changes the facial expression, and the feature points may be registered. Alternatively, the user may select an arbitrary face from already captured image data, and the feature points of the face may be registered as the feature data of the person and expression specified by the user.

  Next, the operation at the time of personal authentication will be described using the flowchart of FIG. This flowchart is started when image data is input to the image input unit 1. If the image input unit 1 is a camera, this image data corresponds to image data photographed by this camera or image data read from a recording medium. If the image input unit 1 is a personal computer, it corresponds to an image read from a recording medium or image data received via a network. This image data may be a still image or a moving image. If it is a moving image, personal authentication is continuously performed at frame intervals corresponding to the time required for personal authentication. In the following embodiment, a description will be given assuming that moving image data is input to the image input unit 1.

  In step S <b> 101, the face detection unit 2 receives one frame of image data of a moving image from the image input unit 1 and detects a human face. In step S102, the face detection unit 2 proceeds to step S103 if one or more faces can be detected, and proceeds to step S111 if no face can be detected. In step S111, it is determined whether image data of another frame exists in the image data input to the image input unit 1, and if it exists, the image data is updated in step S112. Then, returning to step S101, face detection is performed on the image data updated by the face detection unit 2.

  In step S103, the same person determination unit 3 receives the detection result of the face detected by the face detection unit 2, and which face and which face can be regarded as the same person's face among the faces detected in different frames. Determine. When there are a plurality of persons in the image data, it is necessary to distinguish each face and obtain feature data that is the above-described expressionless expression reference for each face. Therefore, the same person determination unit 3 compares the center position and size of each face detected in each frame, and the distance between the center positions of the faces and the sum of changes in size are the most between consecutive frames. A small object is estimated to be the face of the same person. However, even if this sum is the smallest, if the value is not within the preset threshold range, it is determined that the same person does not exist.

  The same person determination unit 3 may estimate only the same person by comparing only one of the position and size of each face, and the person with the most similar detected face brightness and color information is the same person. It may be estimated. In this way, by looking at the correlation between the faces detected in the two frames, it is possible to determine whether or not the faces detected in the two frames are the same person. If the image data input to the image input unit 1 is a still image, this step S103 is omitted.

  In step S104, the face detection unit 2 obtains feature data including coordinates of feature points as shown in FIGS. 6 and 7 from each detected face. In step S <b> 105, the facial expression determination unit 4 determines whether each face has no expression using the feature data obtained by the face detection unit 2 and the determination result obtained by the same person determination unit 3. To do. In the present embodiment, the facial expression determination unit 4 determines that the face at that time is expressionless when the amount of change in mouth shape does not exceed a predetermined threshold for a predetermined period, and this expressionless feature. It is assumed that smiles are determined based on the data. Specifically, the facial expression determination unit 4 accumulates feature data of faces determined to be the same person, and observes relative changes in the feature data to determine whether or not the face has no expression. When the expressionless determination unit 4 can determine the expressionless face, the expression determination unit 4 calculates feature data as a reference for expressionlessness from the expressionless face. If the feature data is not sufficiently accumulated and the facial expression determination unit 4 has not been able to calculate feature data serving as a reference for expressionlessness (NO in step S106), S111 and S112 are performed in order to accumulate facial feature data. Then, the process returns to step S101 again.

  If it is determined in step S105 that feature data serving as a reference for expressionlessness is obtained (YES in step S106), facial expression determination unit 4 proceeds to step S107. It should be noted that the facial expression determination unit 4 may skip the process of step S105 until the face for which feature data serving as a reference for expressionlessness has been calculated once in step S105 is lost. Alternatively, even after the feature data that is the reference for expressionlessness is calculated, the facial expression determination unit 4 updates the feature data that becomes the reference for expressionlessness every time new feature data is calculated in step S104. It may be calculated and updated.

  In step S107, the facial expression determination unit 4 compares the latest feature data obtained in step S104 with the feature data serving as a reference for the expressionless expression calculated in step S105 for each face. The facial expression indicated by the feature data is obtained. In step S <b> 108, the database selection unit 5 selects and reads all personal authentication feature data corresponding to the facial expression determination result output from the facial expression determination unit 4 from the personal database unit 8. For example, if the facial expression determination unit 4 determines that a smile is made in step S107, the database selection unit 5 selects all the persons with smiles from the feature data in the personal database unit 8 shown in FIG. Feature data A-2, B-2, and C-2 are selected and read. When feature data of a large number of persons are registered in advance, the database selection unit 5 may select only the feature data included in a specific person or a group of persons designated in advance by the user.

In step S109, for each face, the authentication unit 6 determines who the face is based on the sum of absolute values S of the latest feature data for which facial expression determination has been performed and the feature data for personal authentication corresponding to the facial expression. Identify In step S110, the result output unit 7 receives the personal authentication result from the authentication unit 6, and displays the result superimposed on the image generated from the image data to be authenticated. Of course, the display of the authentication result is not limited to such a display method, and various methods are conceivable. When personal authentication is performed for a plurality of faces at the same time, it is desirable to clarify which face each authentication result is for.
If there is still an image of a frame that should be subject to face detection in the image data input to the image input unit 1 (NO in step S111), the process returns to step S101 via step S112. Otherwise (YES in step S111), this flowchart ends.

  According to the above embodiment, even when the facial expression changes, the facial expression is detected and the personal authentication is performed using the corresponding feature data. Therefore, the personal authentication can be performed with high accuracy.

  [Second Embodiment] Next, as a second embodiment, a case where feature data corresponding to the facial expression detected in the first embodiment is not registered in the personal database unit 8 will be described. The personal authentication device of this embodiment additionally registers in the personal database unit 8 feature data of a facial expression different from the facial expression at the time of authentication, using the personal authentication result of the image data of another frame.

  FIG. 4 shows the configuration of the personal authentication apparatus of the second embodiment, and a feature data registration unit 9 is added to the configuration of FIG. For example, it is assumed that feature data for personal authentication corresponding to a person's facial expression and eye-opening is registered in the personal database unit 8, but feature data for personal authentication corresponding to a smile is not registered. . When it is determined that the person has smiled, the feature data registration unit 9 causes the personal database unit 8 to register the feature data when the person is smiling as feature data for personal authentication corresponding to the smile. For the other components, the same elements as those in FIG.

  The operation of the personal authentication device of this embodiment will be described with reference to FIG. The processes in steps S113 to S115 in FIG. 5 are different from those in FIG. 3 according to the first embodiment. Of the processes in FIG. 5, processes denoted by the same reference numerals as those in FIG. 3 are the same as the processes in FIG. In steps S101 to S107, the face detection unit 2 detects a face from each frame, obtains feature data of the face, and the same person determination unit 3 determines which face among the faces detected in each frame. And which face is the same person. Then, the facial expression determination unit 4 determines the facial expression of the face for which feature data serving as a reference for expressionlessness is obtained.

  In step S113, the feature data registration unit 9 determines whether or not the face whose facial expression has been determined in step S107 has already been personally authenticated. Whether or not the face has already been personally authenticated can be determined from the previous authentication result by the authentication unit 6 and the previous determination result of the same person by the same person determination unit 3. That is, the same person determination unit 3 continues the face for the face personally authenticated with any expression in the previous frame by the method described in step S103 of the first embodiment. It is determined whether tracking is possible.

  If the face whose facial expression is determined in step S107 is a face that has not yet been personally authenticated (YES in step S113), the process proceeds to step S108, and the same processing as in the first embodiment is performed. If the facial expression determined in step S107 is a face that has already been personally authenticated (NO in step S113), the process proceeds to step S114. In step S114, the feature data registration unit 9 determines whether or not the feature data for personal authentication of the person corresponding to the facial expression determined in step S107 is already registered in the personal database unit 8.

  If the feature data for personal authentication of the person corresponding to the facial expression determined in step S107 is already registered in the personal database unit 8 (NO in step S114), the process proceeds to step S108, and the first The same processing as in the embodiment is performed. If the feature data for personal authentication of the person corresponding to the facial expression determined in step S107 is not yet registered in the personal database unit 8 (YES in step S114), the process proceeds to step S115. In step S115, the feature data registration unit 9 registers the feature data obtained from the face of the facial expression in the personal database unit 8 as feature data for personal authentication corresponding to the facial expression of the person.

  To explain with a specific example, it is assumed that the feature data of Mr. B's expressionless expression has already been registered in the personal database unit 8, but the feature data of smile has not been registered yet. It is assumed that it is already determined that the face is Mr. B based on the authentication result using the expressionless feature data. Here, in the image data of the new frame, it is assumed that the facial expression determination unit 4 determines that the face determined to be Mr. B by the same person determination unit 3 is a smile. At this time, the feature data registration unit 9 newly registers the feature data of the face determined to be a smile in the personal database unit 8 as feature data for personal authentication of Mr. B's smile. When the registration is completed, the personal authentication device proceeds to step S112 and performs the same processing as in the first embodiment.

  It should be noted that feature data for each facial expression of the same person is extracted and stored, and when the individual is authenticated, feature data corresponding to an unregistered facial expression is stored in the stored feature data. You may make it register as characteristic data for authentication. In step S108, if the database selection unit 5 determines that the feature data for personal authentication corresponding to the specific facial expression is small, the result output unit 7 may notify the fact. . When the result output unit 7 shows this notification to the user, it is possible to encourage the user to accumulate feature data for personal authentication corresponding to the missing facial expression.

  In step S109, the authentication unit 6 may comprehensively authenticate the face from not only the personal authentication result for one facial expression but also the personal authentication results for a plurality of facial expressions. For example, the authentication unit 6 picks up the smile feature data in a frame in which a facial expression is determined to be a smile in the video, a feature data of no expression in a frame in which the facial expression is determined to be no expression, and a star in a frame in which the facial expression is determined to be Personal authentication is performed using the feature data. As a result, it is only necessary to select a person who is determined to have the highest degree of matching. As a result, when the authentication results of a plurality of face regions match, it is possible to improve the authentication accuracy by using the final authentication result.

  Here, when personal authentication is performed on the same person in the moving image, there is a possibility that the authentication result is different every time the facial expression changes. At this time, the smile has a large degree of unevenness in the degree of opening of the eyes and the way in which both ends of the mouth rise, and the ratio of matching with the feature data for personal authentication is lower than that of the expressionless face. Therefore, the authentication unit 6 may obtain a final authentication result by statistically calculating the authentication results obtained in time series after reducing the weight of the personal authentication result when smiling with a smaller weight than when there is no expression. .

  In addition, when the authentication result is different, such as Mr. A when there is no expression but Mr. B when the eyes are closed, even though they are the same person, the personal authentication of the eyes using this is different. Feature data may be updated. In such a case, the feature data registration unit 9 updates the feature data for personal authentication of the eyes of Mr. A in the personal database unit 8 using the feature data of the eyes that have been determined to be Mr. B. You may make it do.

  According to the above embodiment, feature data that is not registered in advance is automatically added, so that it is possible to perform personal authentication with higher accuracy.

  [Other Embodiments] The present invention includes a case where the computer program for realizing the functions of the above-described embodiments is achieved by supplying the system or apparatus directly or remotely. In that case, a computer such as a system reads and executes the computer program. Therefore, in order to realize the functional processing of the present invention with a computer, the computer program itself installed in the computer also implements the present invention. In that case, as long as it has the function of a program, it may be in the form of object code, a program executed by an interpreter, script data supplied to the OS, or the like.

  Examples of the recording medium (storage medium) for supplying the program include a flexible disk, a hard disk, an optical disk, and a magneto-optical disk. In addition, there are MO, CD-ROM, CD-R, CD-RW, magnetic tape, nonvolatile memory card, ROM, DVD (DVD-ROM, DVD-R) and the like. As another program supply method, a client computer browser can be used to connect to a homepage on the Internet, and the computer program itself of the present invention can be downloaded from the homepage. It can also be supplied by downloading a compressed file including an automatic installation function to a recording medium such as a hard disk. It can also be realized by dividing the computer program constituting the program of the present invention into a plurality of files and downloading each file from a different homepage. That is, a WWW server that allows a plurality of users to download a program file for realizing the functional processing of the present invention on a computer is also included in the present invention.

  In addition, the program of the present invention is encrypted, stored in a storage medium such as a CD-ROM, distributed to users, and a user who satisfies predetermined conditions downloads key information for decryption from a homepage via the Internet. You can also. In this case, the downloaded key information is used to execute the encrypted program and install it on the computer. In addition to the functions of the above-described embodiments being realized by the computer executing the read program, the OS or the like operating on the computer based on the instruction of the program may be a part of the actual processing. Alternatively, it can be realized by performing all of them. Furthermore, after the program read from the recording medium is written in the memory of a function expansion board inserted into the computer or a function expansion unit connected to the computer, the CPU of the board or the like performs a part of the actual processing. Alternatively, it can be realized by performing all of them.

Claims (10)

Facial expression determination means for determining facial expression from a face area of a person included in input image data of a plurality of frames;
A determination unit that compares information of a plurality of face areas included in the image data of the plurality of frames and determines a face area of the same person among the plurality of face areas;
Selecting means for selecting feature data corresponding to the facial expression determined by the facial expression determination means from storage means in which characteristic data for authenticating a plurality of facial expressions and individuals is associated;
Authentication means for authenticating an individual by comparing the feature data of the face area whose facial expression is determined by the facial expression determination means with the feature data selected by the selection means;
Of the face areas included in the image data of the new frame, from the face area determined to be the same person as the face area for which the individual was authenticated in the image data of the previous frame, the personal information not stored in the storage means A personal authentication comprising: a registration unit that stores the feature data in the storage unit as feature data corresponding to another facial expression of the individual when the feature data corresponding to the different facial expression is detected apparatus.   The personal authentication apparatus according to claim 1, wherein the facial expression determination unit calculates a change amount of feature data from image data of a plurality of consecutive frames, and determines a facial expression based on the change amount.   2. The authentication unit according to claim 1, wherein each of the plurality of image data used by the facial expression determination unit for facial expression determination authenticates an individual and determines a final authentication result based on each authentication result. Or the personal authentication apparatus of 2.   The authentication means reduces the weight for the authentication result in the face area determined to be a smile by the facial expression determination means to be smaller than the weight for the authentication result for the face area determined to be another facial expression. The personal authentication device according to claim 3, wherein the determination is made.   The personal authentication apparatus according to claim 4, wherein the different facial expression is an expressionless expression.   6. The apparatus according to claim 1, further comprising means for notifying that feature data corresponding to the facial expression selected by the selection means does not exist when the feature data does not exist in the storage means. Personal authentication device. Expression determination means for determining an expression from a face area of a person included in the input image data of the plurality of frames and information on the plurality of face areas included in the image data of the plurality of frames are compared to A determination unit for determining the face area of the same person, and a personal authentication method by a personal authentication device,
A selection step of selecting feature data corresponding to the facial expression determined by the facial expression determination unit from a storage unit in which characteristic data for authenticating a plurality of facial expressions and individuals is associated and stored;
An authentication step of authenticating an individual by comparing the feature data of the facial region whose facial expression is determined by the facial expression determination means with the feature data selected by the selection step;
Of the face areas included in the image data of the new frame, from the face area determined to be the same person as the face area for which the individual was authenticated in the image data of the previous frame, the personal information not stored in the storage means And a registration step of storing the feature data in the storage means as feature data corresponding to another facial expression of the individual when feature data corresponding to another facial expression is detected. Method. A camera for taking images,
Facial expression determination means for determining facial expression from a face area of a person included in image data of a plurality of captured frames;
A determination unit that compares information of a plurality of face areas included in the image data of the plurality of frames and determines a face area of the same person among the plurality of face areas;
Selecting means for selecting feature data corresponding to the facial expression determined by the facial expression determination means from storage means in which characteristic data for authenticating a plurality of facial expressions and individuals is associated;
Authentication means for authenticating an individual by comparing the feature data of the face area whose facial expression is determined by the facial expression determination means with the feature data selected by the selection means;
Of the face areas included in the image data of the new frame, from the face area determined to be the same person as the face area for which the individual was authenticated in the image data of the previous frame, the personal information not stored in the storage means And a registration unit that stores the feature data in the storage unit as feature data corresponding to another facial expression of the individual when feature data corresponding to another facial expression is detected.   A program for causing a computer to execute the personal authentication method according to claim 7.   A computer-readable storage medium storing a program for causing a computer to execute the personal authentication method according to claim 7.

Images