JP4185657B2 - Biometric information collating device, biometric information collating terminal, and biometric information collating method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a biometric information collation apparatus, a biometric information collation terminal, and a biometric information collation method for collating and confirming biometric information that is raw information unique to a person such as face information, fingerprint information, and voiceprint information.
[0002]
[Prior art]
In general, the collation device for collating biometric information in the above example collates a face image of a person to be collated acquired using a camera and a registered image registered in advance to confirm the person to be collated. .
[0003]
In this case, if the person to be collated closes his eyes, his / her mouth changes his facial expression, or his face changes suddenly, it is different from the registered image. In this case, a re-verification operation is performed to verify the person himself / herself.
[0004]
However, if the person cannot be verified even after performing this re-verification operation, use a key input operation to enter the password with the numeric keypad, or use an authentication card to check the authentication by reading the data on the magnetic card or IC card. Authenticate using the alternative authentication input means.
[0005]
[Problems to be solved by the invention]
However, in this case, if a malicious person knows the PIN, or if an unauthorized person obtains an authentication card from another person, an alternative authentication input method that uses a numeric keypad or authentication card to intentionally fail to match the face image There is a risk that fraudulent input may be made, and in this fraud authentication, there is no evidence of the face image of the person targeted for fraud verification because the face is hidden, etc. It was.
[0006]
Therefore, the present invention provides a biometric information matching apparatus, a biometric information matching, which can improve the fraud prevention effect by using the alternative authentication input means only when the evidence of the person to be matched remains in the use of matching, and improve the security. An object is to provide a terminal and a biometric information matching method.
[0007]
[Means for Solving the Problems]
This invention acquires the biometric information input by the authentication input means which permits the input of biometric information of the person to be verified, and whether or not the acquired biometric information matches the verification data stored in the storage means in advance. Biometric information comprising: a determination unit that determines whether to collate; and a storage unit that stores the input biological information as evidence , and the determination unit can identify the person to be verified based on the input biological information If it is determined that the input biometric information matches the collation data stored in advance in the storage means, the person to be collated is determined to be authentic, and the determination means Is a case where it is determined that biometric information that can identify the person to be collated can be acquired based on the input biometric information, and the input biometric information is stored in the storage unit in advance. When it is determined that the data does not match, the input biometric information is stored in the storage unit as evidence, and an authentication input by an authentication input unit different from the authentication input unit is allowed to the person to be collated. When it is determined that the biometric information that can identify the person to be verified cannot be acquired based on the input biometric information, authentication input by an authentication input means different from the authentication input means is not permitted to the person to be verified. It is characterized by that.
[0008]
As a result, since the biometric information that can identify the verification target person can be acquired, the verification target person's own evidence remains in the storage unit, and therefore, it is determined that it does not match the verification data stored in the storage unit in advance. In this case, the use of other authentication input means can be permitted to newly authenticate the person to be verified. On the other hand, if biometric information that can identify the person to be collated cannot be acquired, the person to be collated cannot be confirmed, and no evidence remains at that time. Actions can be prevented in advance.
[0009]
Therefore, even if a malicious person intentionally fails to collate face images and attempts other authentication input acts illegally using a numeric keypad or an authentication card, it can be rejected because there is no evidence. Further, even if the fraudster tries to authenticate using other authentication input means, since the evidence of the biometric information remains in the storage means, the fraudulent person can be reliably identified.
[0010]
In this way, since it has a check function to monitor the biometric information that is evidence of all the people who are used for verification, it is possible to prevent fraud authentication from happening because alternative verification can not be performed unless evidence of the person to be verified remains. As a result, security can be improved.
[0011]
In another invention, the biometric information inputted by the authentication input means that allows the biometric information of the person to be collated is transmitted to the center apparatus, and the biometric information transmitted from the center apparatus is stored in advance. A biometric information collating terminal that receives a collation determination result indicating whether or not the data matches the data and displays the collation determination result can be specified by the center device based on the transmitted biometric information. If it is determined that the information has been acquired and if it is determined that the verification result for the transmitted biometric information matches, the verification target person is determined to be authentic, and the center device Based on the transmitted biometric information, it is determined that biometric information that can identify the person to be collated has been acquired, and the collation determination result for the transmitted biometric information is When it is determined not to do so, the input biometric information is stored as evidence in a storage unit provided in the center device, and authentication input by another authentication input unit is allowed to the person to be verified, and the center device When it is determined that biometric information that can identify the person to be collated cannot be acquired based on the transmitted biometric information, authentication input by other authentication input means is not permitted to the person to be collated. And
[0012]
In this case, when the biometric information that can identify the person to be collated cannot be acquired, the fact is sent to the center device to invalidate other authentication input actions. At this time, the collation judgment result of the person to be collated is displayed. Since it can be displayed on the biometric information collating terminal and the collation target person can be informed that the biometric information of the person cannot be acquired, the fraud deterrent effect can be enhanced.
[0013]
Conversely, even if it is determined that the verification determination results do not match and an unauthorized person is authenticated based on an authentication input by another authentication input means, biometric information as evidence is stored in the storage means provided in the center apparatus. Since it is left behind, it is possible to identify the fraudster. Therefore, security can be improved.
[0014]
In another invention, the authentication input means for acquiring the biometric information input while allowing the biometric information of the person to be collated to be input, the storage means in which the biometric information of the person to be collated is stored in advance as collation data, A biometric information collation apparatus comprising: a determination unit that determines whether or not the acquired biometric information matches the collation data stored in advance in the storage unit; and a storage unit that stores the input biometric information as evidence The biometric information collation method using the information processing method is the case where the determination means determines that biometric information that can identify the person to be collated can be acquired based on the input biometric information, and the input is performed. and the biometric information is determined to match the verification data stored in advance in the storage means, determines that the authentic the collating subject, the determining means, said verification based on the inputted biometric information When it is determined that the biometric information that can identify the elephant has been acquired, and when it is determined that the input biometric information does not match the collation data stored in the storage unit in advance, the input biometric information The information is stored as evidence in the storage unit, the authentication input by another authentication input unit is allowed to the verification target person, and the determination unit can identify the verification target person based on the input biometric information When it is determined that the biometric information could not be acquired, the biometric information collating method does not allow the person to be collated to input authentication by other authentication input means.
[0015]
In this case, since the evidence of the person to be collated remains in the storage means when the biometric information that can identify the person to be collated can be obtained, if it is determined that the collation does not match, other authentication input means can be used. Continue verification. On the other hand, if biometric information that can identify the person to be collated cannot be acquired, no evidence of the person to be collated remains, and thereafter, the authentication can be disabled by restricting the acceptance of other authentication input means. Can be secured.
[0016]
Here, the biometric information is mainly face information of the person to be collated, and also means information unique to the person such as fingerprint information and voiceprint information.
The other authentication input means means alternative authentication information other than the already acquired biometric information. For example, confirmation of a personal identification number using a numeric keypad, confirmation of ID information recorded on an ID card, fingerprint or voiceprint Including confirmation.
[0017]
In particular, the biometric information is used as face image information, and it is determined whether or not the acquired biometric information is biometric information that can identify a person to be collated. By performing whether or not there is, the face image information of the person to be collated serves as evidence for identifying the person to be collated as it is, so that the fraudulent person can be reliably identified and the fraud deterrent effect can be further enhanced.
[0018]
DETAILED DESCRIPTION OF THE INVENTION
An embodiment of the present invention will be described below in detail with reference to the drawings.
The drawing shows a face image collation system for photographing and collating a human face. In FIG. 1, a face image collation apparatus 11 is installed on the authentication operation side of the face image collation system, and a center apparatus 12 is installed on the authentication monitoring side. A monitor 13 and a keyboard 14 are installed.
[0019]
The above-mentioned face image collation device 11 is installed for opening / closing management of a door to be used, for example, and is attached to a person's face height position on the door surface with a collation data acquisition function and an authentication data acquisition function. An illumination device 15 for illuminating the face is installed in the upper part, a camera 16, a human detection sensor 17, a collation result display LED 18, and a collation start button 19 are arranged in the middle part. A (numerical key) 20 and a card reader 21 are arranged.
[0020]
The illumination device 15 arranges a plurality of illumination LEDs and illuminates toward the front face height position. Thereby, the facial image acquisition performance of the camera 16 is enhanced by clearly illuminating the uneven state of the face of the person approaching the camera 16.
[0021]
The camera 16 takes a face image using a CCD camera or the like. When a person approaches the front of the camera 16, the person detection sensor 17 such as a reflective infrared sensor detects that the person has approached and further collates. When the start button 19 is pressed, it is determined that the collation is requested by the person to be collated, and shooting of the camera 16 is started, and collation is started based on the acquired image. At this time, the collation result of the collation target person that has been collated is displayed and output on the collation result display LED 18. The collation result display LED 18 lights a blue or red determination result on one of the two LEDs for each color so that the collation suitability can be recognized at a glance.
[0022]
When using the above-described verification, when a face image is acquired by the camera 16, if it is confirmed that a clear face image that can verify the verification target person has been acquired, the image remains as evidence. Even if it is determined, other authentication input means are accepted and allowed to continue the collation operation. On the other hand, when a face image is acquired, if the person to be verified cannot identify the person due to actions such as closing his eyes, opening his mouth or hiding his face, the image has no value as evidence. Because the image has insufficient evidence, the security is improved by rejecting acceptance of other authentication input means at that time in order to restrict the continuation of verification.
[0023]
At this time, the normal authentication input means is an operation for checking the face image acquired based on the pressing operation of the collation start button 19 described above against the registered data, and the face image is not used as the other authentication input means. One of the personal identification key 20 and the card reader 21 is used, and the personal identification key 20 and the card reader 21 have an independent authentication function, and a personal identification number for identifying a given person is assigned to the personal identification key 20. The person to be verified performs an input operation using. When the card reader 21 is used, authentication is performed by reading ID data of a magnetic card or IC card in which ID data for specifying the person to be verified is stored.
[0024]
The verification data acquired in this way is guided to the center device 12 described later on the authentication monitoring side via the connection cable 22 or wirelessly, and the suitability of the verification data is determined here. Moreover, the entrance / exit utilization state is monitored by a monitor 13 connected to the center device 12, and the keyboard 14 is used to permit data input and door opening / closing control by the supervisor.
[0025]
In the center device 12 described above, face information of each registrant determined in advance is registered, and the face image information of the person to be collated photographed by the camera 16 and the registration information of the center device 12 are collated to collate. Determine whether the subject matches or does not match.
[0026]
FIG. 2 shows a control circuit block diagram of the face image matching system, which is controlled and managed by the center device 12 that controls each device provided in the face image matching device 11.
The CPU 23 of the center device 12 controls each circuit device in accordance with the stored program and stores the control data in the memory 24.
[0027]
The image capture 25 acquires an image captured by the camera 16 and stores it in the memory 24. The acquired image is displayed on the monitor 13 for monitoring via the display control unit 26.
[0028]
The first input / output control unit 27 controls input / output data of the illumination device 15, the human detection sensor 17, the verification result display LED 18, and the verification start button 19 provided in the face image verification device 11, and the second input / output control unit 28. Controls input from the authentication monitoring keyboard 14.
[0029]
The RS232C communication unit 29 has a communication connection function for transmitting input data from the card reader 21 and the personal identification key 20 to the CPU 23.
[0030]
The registration file 30 is provided as a database file at the time of face image registration, and data of face image feature values of each registrant is stored therein.
[0031]
Then, the CPU 23 collates the face image information of the person to be collated photographed by the camera 16 with the registered information registered in advance, and performs entrance / exit permission or room entry / exit regulation based on the collated determination result.
[0032]
FIG. 3 is an explanatory diagram of the control operation of the center device 12. The captured image is collated with face authentication data recorded in advance in the face authentication data file 32 in the face collation authentication unit 31, and the authentication result and The image is recorded in the usage record data file 34 of the usage recording unit 33.
[0033]
At this time, when the face image of the person to be collated is acquired, the use record determination unit 35 determines whether or not the image includes an appropriate face of the person to be collated and can identify the person to be collated If the image information is satisfied, even if it is determined that the verification is impossible, the input operation of the alternative authentication means 36 using the personal identification key 20 or the card reader 21 can be accepted and allowed to be authenticated. Processing can be completed.
[0034]
At the time of this alternative authentication, the authentication data in the alternative authentication data file 37 recorded in advance is collated, the authentication result is recorded in the usage record data file 34 of the usage recording unit 33, and the determination result output unit 38 is notified. Then, the determination result output unit 38 outputs any of the alternative authentication results from the face matching authentication unit 31 to the matching result display LED 18 of the face image matching device 11 and the monitor 13 of the center device 12.
[0035]
On the other hand, if it is determined that the face image of the person to be collated cannot be confirmed at the stage when the face matching authentication unit 31 acquires the face image, for example, the face is hidden or the facial expression is more than usual. In the case of an indistinguishable face image in which the person cannot be identified due to a large change, since there is no clear evidence of the face image, the use of the alternative authentication means 36 using the above-described password 20 or card reader 21 is restricted. Therefore, it is possible to prevent unauthorized authentication.
[0036]
Therefore, even if a malicious person intentionally fails to match the face image and attempts to cheat using another person's PIN or other person's card, it is accepted because there is no evidence of the face image that identifies the person. You can refuse. In addition, when a fraudulent person can use other alternative authentication means 36, since the evidence of the face image remains, the fraudulent person can be reliably identified.
[0037]
As a result, when the face image cannot be acquired, the fact is transmitted to the center device 12 to invalidate the alternative authentication action. At this time, the collation determination result of the collation target person is displayed as the collation result of the face image collation device 11. Since it can be displayed on the LED 18 and notified to the person to be collated that the face image of the person could not be acquired, and in this case, operation of high collation use such as re-acquisition of the face image can be achieved. The fraud deterrence effect can be increased accurately.
[0038]
A collation processing operation using the face image collation system configured as described above will be described with reference to the flowcharts of FIGS.
The face image collation device 11 is installed on the door of the room, and the person to be collated starts in front of the camera 16 while the human detection sensor 17 detects and confirms that a person using the entrance / exit has approached the door. When the button 19 is pressed (step n1),
The CPU 23 uses the camera 16 to acquire the face image of the person to be collated immediately after the collation start button 19 is pressed. The face image data acquired by the camera 16 is stored in the memory 24 (step n2).
[0039]
In this case, the CPU 23 searches the position of the face from the eyes, nose, mouth, etc., which are the features, and cuts out a part (step n3),
If a portion having a high face-likeness is cut out, facial feature quantities such as eyes, nose and mouth, which are features at the time of collation, are extracted (steps n4 to n5).
[0040]
At this time, if it is confirmed that the extracted eyes, nose, mouth and the like are all high (step n6),
The feature amount of the face image is obtained, and the similarity is determined by comparing and collating the feature amount of the acquired face image with the feature amount of the face image registered in advance (steps n7 to n8).
[0041]
If the results of the collation determination match, the collation result display LED 18 is lit to indicate that it matches, and the entry / exit door is unlocked to allow entry / exit of the person to be collated (steps n9 to n10).
[0042]
On the other hand, if the face-like portion suitable for collation cannot be extracted in step n4, for example, the collation target person is determined to be a collation failure image by closing his eyes, opening his mouth, or turning his gaze. In this case, since the person cannot be identified from the acquired verification failure image and does not remain as evidence of the verification target person himself / herself, even if there is a key input operation to the personal identification key 20, the alternative authentication used for the card by the card reader 21 Even if the operation is performed, the acceptance is rejected, an error display by lighting or blinking is performed on the collation result display LED 18, and the error occurrence display guidance is also performed on the monitor 13 on the center device 12 side (steps n11 to n13).
[0043]
On the other hand, if the value of the feature quantity such as eyes, nose, mouth, etc. is low in step n6, the CPU 23 determines that the result is an inappropriate determination result, and displays on the verification result display LED 18 that the authentication has failed (step n14).
[0044]
By the way, when the face image of the person to be collated obtained immediately after pressing the collation start button 19 is collated in the above step n9, the feature amount of the obtained face image and the face image stored in the registration file 30 in advance are stored. Even if the CPU 23 compares and determines the feature amount, and it is determined that the comparison determination result is not possible at this time, when the face image is acquired in advance, the verification target person himself can be confirmed from the facial appearance. Therefore, if it is subsequently determined that collation is not possible, use of another authentication input means can be permitted to newly authenticate the person to be collated. On the other hand, if an appropriate face image cannot be obtained, the person to be verified cannot be confirmed, and no evidence remains at that time, so that acceptance of other authentication input means using a personal identification key or a card reader is rejected.
[0045]
The CPU 23 can perform an alternative authentication operation using the personal identification key 20 or the card reader 21 (steps n15 to n16).
[0046]
When authenticated, the verification result display LED 18 is lit to indicate that the authentication was successful, and the entrance / exit door is unlocked to allow entry / exit.
However, if the data does not match even if the personal identification key 20 or the card reader 21 is used and the data is not authenticated, an improper determination result indicating that the collation processing is impossible is displayed on the collation result display LED 18 to restrict entry / exit (step n17). ~ N18).
[0047]
As described above, if the face image can be clearly acquired, the person can be identified, so that the subsequent verification can be continued, and if the face image cannot be clearly acquired, the verification target person's own confirmation Since no evidence remains at that time, switching use of other authentication input means using a secret key or a card reader is refused.
[0048]
Therefore, when there is no evidence of a face image, it is possible to refuse acceptance of other authentication input acts, and even if an unauthorized person attempts to authenticate using other authentication input means after acquiring a proper face image. Since the evidence of the face image remains, it is possible to reliably identify the fraudster. As described above, since it has a check function for monitoring face images as evidence of all persons used for verification, it is possible to prevent an unauthorized authentication action and to improve security accordingly.
[0049]
In correspondence between this invention and the configuration of the above-described embodiment,
The biometric information matching device of this invention corresponds to the face image matching device 11 and the center device 12 of the embodiment,
Similarly,
The storage means corresponds to the memory 24,
The biometric information matching terminal corresponds to the face image matching device 11,
The storage means corresponds to the registration file 30,
The determination means corresponds to the center device 12,
Different authentication input means correspond to the password 20 and the card reader 21,
The biometric information authentication input means corresponds to the camera 16,
The present invention can be applied based on the technical idea shown in the claims, and is not limited to the configuration of the above-described embodiment.
[0050]
For example, in the above-described embodiment, the face image data and authentication data input side is separated from the center device 12 side that authenticates them, but the present invention is not limited to this. It is also possible to have a single structure.
[0051]
In addition, although a face image is shown as an example of biometric information, a person to be verified can be determined based on fingerprint information or voiceprint information.
[0052]
【The invention's effect】
According to the present invention, since the evidence for each verification target person can be secured at the stage where the biometric information that can identify the verification target person is input to the biometric information authentication input means, if the evidence can be confirmed, the verification use is continued. It is possible to use other authentication input means. If evidence cannot be secured, it is possible to prevent fraudulent acts by rejecting authentication inputs from other authentication input means.
[Brief description of the drawings]
FIG. 1 is a schematic configuration diagram of a face image matching system according to the present invention.
FIG. 2 is a block diagram of a control circuit of the face image matching system according to the present invention.
FIG. 3 is an explanatory diagram of a control operation of the center device of the present invention.
FIG. 4 is a flowchart showing the collation processing operation of the face image collation system of the present invention.
FIG. 5 is a flowchart following FIG. 4;
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 11 ... Face image collation apparatus 12 ... Center apparatus 16 ... Camera 18 ... Collation result display LED
20 ... PIN key 21 ... Card reader 23 ... CPU
24 ... Memory 31 ... Face verification authentication unit 35 ... Usage record determination unit 36 ... Alternative authentication means

Claims (4)

The biometric information input by the authentication input means that allows the input of biometric information of the person to be collated is acquired, and it is verified whether or not the acquired biometric information matches the collation data stored in the storage means in advance. Determination means, and storage means for storing the input biological information as evidence,
The determination means determines that the biological information that can identify the person to be collated can be acquired based on the input biological information, and the input biological information is stored in the storage means in advance. If it is determined that the verification data matches the verification data, the verification target person is determined to be authentic,
The determination means determines that the biological information that can identify the person to be collated can be acquired based on the input biological information, and the input biological information is stored in the storage means in advance. When it is determined that it does not match the verification data, the input biometric information is stored in the storage unit as evidence, and an authentication input by an authentication input unit different from the authentication input unit is allowed to the verification target person,
If the determination means determines that biometric information that can identify the person to be verified cannot be acquired based on the input biometric information, an authentication input by an authentication input means different from the authentication input means is to be verified. A biometric information collating apparatus characterized by being not allowed to a person. The biological information is face image information,
The determination as to whether or not the acquired biometric information is biometric information that can identify the person to be collated is performed based on whether or not the feature quantity based on the facial part based on the face image information is greater than or equal to a predetermined value. The biometric information collating apparatus according to claim 1, wherein The biometric information input by the authentication input means that allows the input of biometric information of the person to be collated is transmitted to the center device, and whether or not the biometric information transmitted from the center device matches the collation data stored in advance. A biometric information matching terminal that receives the matching determination result and displays the matching determination result,
When it is determined by the center device that biometric information that can identify the person to be verified can be acquired based on the transmitted biometric information, and the collation determination result for the transmitted biometric information matches. If determined, the person to be verified is determined to be authentic,
The center device determines that biometric information that can identify the person to be collated has been acquired based on the transmitted biometric information, and the collation determination result for the transmitted biometric information does not match When it is determined that the biometric information that has been input is stored as evidence in the storage unit provided in the center device, and authentication input by an authentication input unit different from the authentication input unit is allowed to the verification target person,
When it is determined by the center device that biometric information that can identify the person to be verified cannot be acquired based on the transmitted biometric information, verification input by an authentication input unit different from the authentication input unit is verified. A biometric information collating terminal characterized in that it is not allowed for a subject. Authentication input means for acquiring biometric information input while allowing biometric information of the person to be collated to be input, storage means for preliminarily storing the biometric information of the person to be collated as collation data, and Biometric information collation using a biometric information collating apparatus comprising: a determination unit that determines whether or not the collation data stored in the storage unit is coincident; and a storage unit that stores the input biometric information as evidence A method,
The determination means determines that the biological information that can identify the person to be collated can be acquired based on the input biological information , and the input biological information is stored in the storage means in advance. If it is determined that the verification data matches the verification data, the verification target person is determined to be authentic,
The determination means determines that the biological information that can identify the person to be collated can be acquired based on the input biological information, and the input biological information is stored in the storage means in advance. When it is determined that it does not match the verification data, the input biometric information is stored in the storage unit as evidence, and an authentication input by an authentication input unit different from the authentication input unit is allowed to the verification target person,
If the determination means determines that biometric information that can identify the person to be verified cannot be acquired based on the input biometric information, an authentication input by an authentication input means different from the authentication input means is to be verified. A biometric information collating method characterized in that it is not allowed to a person.

Images