JP2002150296A - Authentication method and authentication device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an authentication device that prevents unrestricted authentication of a third party in individual authentication. SOLUTION: For authentication of a visitor to facilities, a person to be authenticated as a visitor inserts one hand in an authentication box 3 to make a series of actions, for example, showing paper, scissors and stone, and the hand actions are shot by a camera device 6. The shot image is subjected to given image processing in an image processing part 8. A decision part 10 then determines a coefficient of correlation between the image-processed image data and preliminarily registered action image data on the hand of an authentic person, and if finding it equal to or larger than a given threshold value, decides that the person in question is the authentic person related to the action image data. The authentication involves a habit peculiar to the authentic person.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and apparatus for authenticating an individual.

[0002]

2. Description of the Related Art Conventionally, as an authentication device provided at an entrance / exit point of an important facility, there is an authentication device which inputs a password from a numeric keypad.

[0003]

In a contact-type device for inputting a personal identification number from a numeric keypad, a personal identification number is guessed by a third party from the degree of wear and dirt of the key, and when the personal identification number is known, it is simply obtained. However, there is a disadvantage that the authentication can be easily obtained only by pressing the password with the numeric keypad.

Accordingly, an object of the present invention is to provide an authentication method and an authentication device in which a third party is not easily authenticated when performing personal authentication.

[0005]

In order to solve the above-mentioned problems, an authentication method according to the present invention takes an image of a part of the body of a person to be authenticated, and obtains a photographed image of a part of the body of the person to be authenticated. This is a method of authenticating a person to be authenticated by judging the degree of similarity with the motion image.

Further, a first authentication device of the present invention comprises:
An image processing unit for inputting an image photographed by the photographing device and performing image processing, and calculating a similarity between image data obtained by the image processing unit and motion image data of a pre-registered authenticator; And a determination unit for determining whether or not to authenticate the user based on the similarity.

[0007] A second authentication device of the present invention includes:
An image processing unit that inputs an image captured by the image capturing apparatus and performs image processing, and determines a similarity between image data obtained by the image processing unit and motion image data of a pre-registered authenticator. In this case, a decision unit for making a decision by a neural network method is provided.

According to the above-described authentication method and the configuration of each authentication device, since the judgment is made based on the operation of a part of the body of the person to be authenticated, for example, a habit peculiar to the authenticator is interposed. Compared to the case of using, the third party is hardly erroneously authenticated, so that the security at the time of authentication is very high.

[0009]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An authentication method and an authentication device according to a first embodiment of the present invention will be described below with reference to FIG.

[0010] The authentication device according to the first embodiment includes:
For example, it is provided at the entrance and exit of an important facility to determine whether or not a visitor is a person who has been permitted to enter the facility. It is to identify visitors and give authentication.

As shown in FIG. 1, the authentication device 1 is provided at, for example, an entrance of an important facility, and is capable of inserting a hand of a visitor (hereinafter referred to as a person to be authenticated) and detecting the insertion. (For example, a photoelectric tube type is used.)
An illumination box 5 provided in the authentication box 3 and illuminating light by operating the switch unit 4 in response to a hand detection signal from the detector 2;
Camera device (for example, a CCD camera is used) 6 for photographing the hand inserted in the authentication box 3
And an image captured by the camera device 6 and a selection signal (hereinafter, referred to as a trigger signal) from the selection signal output unit 7 to select (determine) an image to be used for authentication, and select the selected image. An image processing unit 8 for performing predetermined image processing on the image, a storage unit 9 storing image data of hand movements of a person permitted to enter (hereinafter referred to as an authenticator), The image data obtained by the processing unit 8 and the image data from the storage unit 9 are input to determine whether the person to be authenticated is the authenticator and, if so, to identify the authenticator. And a judgment unit 10 for performing the judgment.

In the storage unit 9, image data of each operation of the authenticator's hand, for example, each operation such as goo, choki and par, and data of a flow of a series of operations (hereinafter, referred to as a series of operations) are registered. ing.

The output timing of the selection signal from the selection signal output unit 7 is also predetermined. For example, for one operation, 10 trigger signals are output within a short interval (for example, about 4 seconds) (of course, 10
The number of operations is not limited to the number of times, and may be increased or decreased as necessary. Therefore, it is determined whether or not each operation belongs to the authenticator based on the ten captured images. Note that the hand insertion detection signal from the detector 2 is also input to the camera device 6 and the selection signal output unit 7. When the hand is inserted into the authentication box 3, the camera device 6 shoots. It is started and is used as the reference time of the trigger signal output. The selection signal from the selection signal output unit 7 is also input to another component unit, for example, the determination unit 10 as necessary.

Next, an authentication operation (authentication method) by the authentication device will be described. A person to be authenticated, who is an admission to an important facility, inserts his hand into the authentication box 3 provided at the entrance, and performs, for example, goo, choki, and par operations in accordance with a series of pre-registered operations. .

That is, when the person to be authenticated inserts his / her hand into the authentication box 3, the detector 2 detects the hand insertion,
The lighting is applied to the hand by the lighting device 5 via the switch unit 4.

In the authentication box 3, photographing by the camera device 6 is started based on the detection signal of the hand insertion, and a photographed image of the inserted hand (specifically, the back of the hand) is processed by the image processing unit 8. Is input to

In the image processing section 8, an image to be determined is specified by a trigger signal output from the selection signal output section 7, and predetermined image processing is performed on the specified image. For example, processing such as image binarization processing, edge processing, and enlargement / reduction is performed.

Then, the image data subjected to the image processing and the reference image data input to the storage unit 9 are input to the determination unit 10, where the similarity between the two image data is calculated.

For one operation, for example, ten image data are obtained, and a correlation coefficient between each of the image data and the reference image data is obtained. At this time, a weight coefficient is used for the beginning and end of each operation because there is a high possibility that the operation is shifted from the correct operation.

That is, each trigger signal corresponding to 10 sheets
Weighting factor w for the signal_iIs set and the operation
For the beginning and end, the weighting factor w_iIs small
It is set up. For example, trigger 1 to trigger 1
0, the weight coefficient w _iIs 0.3; 0.5;
0.8; 1.0; 1.0; 1.0; 1.0; 0.8;
0.5; set to 0.3.

For example, in the case of a Goo operation, a correlation coefficient is obtained for each of the ten image data, and the above-mentioned weight coefficient w _{i is} added to the correlation coefficient.
Are obtained, and the sum of these values is obtained as an overall evaluation value, and the overall evaluation value is compared with a predetermined threshold value.

If the total evaluation value is equal to or larger than the threshold value, it is determined that the reference image data is applicable, and the authenticator is identified from the identification data added to the image data. In this case, a case where a plurality of persons correspond is specified.

When this determination and specification are performed for all the series of operations, and all the operations are the same authenticator, or when there are a plurality of persons who are involved in the determination of each operation, in all the operations, If there is a corresponding (common) certifier, it is determined whether or not the certifier matches the data of the series of operations registered.

If the sequence of operations is matched, the authenticator is finally determined, and a "suitable" signal is output to the door opening / closing device. Of course, if the total evaluation value is smaller than the threshold value, a signal of "inappropriate (no)" is output, and for example, a display of refusal of entry or a display for urging re-authentication is displayed.

When the person to be authenticated leaves the authentication box 3 after the end of the authentication, the illumination by the lighting device 5 is extinguished and the photographing by the camera device 6 is stopped. Note that history items such as data of the authenticator, the date and time of entry, and the data of denied entry are stored in the storage unit 9.
And will be used for safety management verification at a later date.

By the way, in the above embodiment, 1
Shooting multiple times for one action,
For example, shooting may be performed only once for one operation.

At this time, when performing a series of operations, a predetermined time (for example, about 5 seconds) is provided between each operation, and the time of this operation interval may be used as judgment data. it can.

In this case, the authenticator is informed in advance of the number of seconds until the first operation after inserting the hand into the authentication box 3 and the number of seconds between operations. Shooting is performed. Of course, this photographing instruction is output from the selection signal output unit 7.

Further, in the above-described embodiment, the camera device is operated by detecting with the detector that the person to be authenticated has inserted his / her hand into the authentication box. For example, an alarm device that generates an alarm may be activated, or a monitoring camera that captures an image of the person to be authenticated may be provided to project the person to be authenticated on a monitor device.

As described above, since the judgment is made based on the movement of a part of the hand of the person to be authenticated, a habit (characteristic) unique to the authenticator, such as the shape and movement of the hand of the individual, is interposed. Individuals differ in how they hold their hands, how they open, how they lean, etc., and therefore are less likely to be imitated by others than in the case of using a password as in the past (generally, In other words, it is almost impossible for a third party to be incorrectly authenticated), and the security at the time of authentication is very high. Of course, the degree of authentication can be further increased by increasing the number of hand movements and changing the interval at which the movements are performed.

As for the hand movement for performing the authentication, only the person who is permitted to enter (authenticate) is informed,
Since there are a large number of such combinations due to operation intervals, it is almost impossible that another person who is not permitted to enter is erroneously permitted (it is possible only by chance).

Further, the non-contact type authentication makes it difficult to make unauthorized electrical and mechanical authentication, and it takes time for unauthorized entry even if it is attempted to enter illegally.
This leads to easier discovery of others, ie, suspicious individuals.

Further, at the time of authentication, by inserting a hand into the authentication box, the authentication operation (authentication contents) cannot be known from the outside, so that the security is further ensured.

Next, a second embodiment of the present invention will be described with reference to FIGS. In the first embodiment, the degree of similarity between image data is used as a determination method in the determination unit. In the second embodiment, a neural network method is used.

The second embodiment differs from the first embodiment in the judgment method. Therefore, the following description focuses on the parts related to the judgment method, and also describes the first embodiment. The same components as those described above are denoted by the same reference numerals and description thereof will be omitted.

That is, as shown in FIGS. 2 and 3,
In the image processing unit 8 ', predetermined image processing, for example, binarization processing, edge processing, enlargement / reduction processing, etc., is performed on the captured image input from the camera device 6, and these processings are further performed. As shown in FIG. 3A, the image data is divided into a large number of rectangular (not limited to rectangular, for example, hexagonal) small areas 8a, and Each time an edge is detected, and if an edge exists, data of “1” is
If no edge exists, data “0” is added.

On the other hand, in the judgment section 10 ', as shown in FIG. 3B, an input layer 11 having an input section 11a corresponding to each small area of the image data, and "goo", "choki", An output layer 12 having an output section 12a corresponding to an operation content such as "par" and "other" (other than the authenticator) indicating other than the registered operation, and the input layer 11 and the output layer 12 are connected. A neural network unit 14 including an intermediate layer (hidden layer) 13 having a coupling unit 13a is provided. Incidentally, the connection weight k _i for coupling together the coupling portion 13a and the input unit 11a in advance, are set by learning.

The neural network unit 14 provided in the determination unit 10 'is for specifying a photographed image. Therefore, based on each specified image data, it is determined whether or not the user is an authenticator. Is determined by the first
This is performed in the same manner as in the embodiment. For example, determination of data of a series of operations is performed.

In the case of the second embodiment, the same effects as those of the first embodiment can be obtained. By the way, in each of the above-described embodiments, the hand motion is used for performing the authentication. However, in addition to the hand motion, for example, a gesture using the upper body by gesture, hand gesture, or the like may be used.

In each of the above embodiments, when calculating the similarity between the image data of the authenticated person and the authenticated person,
Although the correlation coefficient is used, the similarity may be obtained by, for example, pattern matching.

[0041]

As described above, according to the authentication method and the authentication apparatus of the present invention, the judgment is made based on the operation of a part of the body of the person to be authenticated. For example, compared to the case where a password is used, a third party is hardly erroneously authenticated, so that the security at the time of authentication is very high.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a schematic configuration of an authentication device according to a first embodiment of the present invention.

FIG. 2 is a block diagram illustrating a schematic configuration of an authentication device according to a second embodiment of the present invention.

FIG. 3 is a diagram illustrating a determination method in a determination unit in an authentication device according to a second embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 Authentication apparatus 2 Detector 3 Authentication box 4 Switch part 5 Illumination tool 6 Camera device 7 Selection signal output part 8 Image processing part 8 'Image processing part 9 Storage part 10 Judgment part 10' Judgment part 11 Input layer 12 Output layer 13 Middle layer 14 Neural net part

 ──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Kazuhisa Hayashi 1-7-89 Minami Kohoku, Suminoe-ku, Osaka-shi, Osaka F-term in Hitachi Zosen Corporation (reference) 4C038 VA07 VB12 VC01 VC05 5B057 AA20 BA02 CA12 CA16 DA06 DB02 DC22 DC36 5L096 BA18 CA02 HA03 HA07 JA11

Claims (4)

[Claims] An authentication of a person to be authenticated is performed by photographing an operation of a part of the body of the person to be authenticated and judging a similarity between the photographed image and a motion image of a previously registered authenticator. An authentication method characterized in that: 2. A photographing apparatus for photographing a part of the body of a person to be authenticated, an image processing section for inputting an image photographed by the photographing apparatus and performing image processing, A calculating unit that calculates a similarity between the registered image data and the motion image data of a pre-registered authenticator, and that determines whether to authenticate the user based on the similarity. apparatus. 3. A photographing device for photographing a part of the body of a person to be authenticated, an image processing unit for inputting an image photographed by the photographing device and performing image processing, An authentication device for determining a similarity between the registered image data and motion image data of a pre-registered authenticator by a neural network method. 4. The authentication device according to claim 2, wherein a hand motion is used as the authentication operation.