JP2005115460A - Device for collating face, and passage controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a face collating device for reliably rejecting spoofing using a facial portrait or a video image, etc., and identifying a face which is not discriminated by face collation with an image processing only. <P>SOLUTION: The face collating device collates the face by using the facial image of a person to be collated, and authenticates the identical person by combining face collation with the facial image of the person to be collated and that with the facial temperature distribution of a human being. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention provides, for example, a face matching device that determines whether or not a person to be verified is a person by comparing a face image input by the person to be verified with dictionary information registered in advance, and the face matching device The present invention relates to a traffic control device that controls the traffic of a passerby (for example, entry into and exit from a security area).

  In general, this type of face collation apparatus obtains a face image of a person to be collated as a collation target using a camera or the like, and dictionary storage means using the person-specific feature information obtained from the obtained face image as dictionary information Is stored (registered) in advance.

At the time of collation, the face image of the person to be collated as a collation target is acquired again using a camera or the like, and the person-specific feature information obtained from the acquired face image is stored in the dictionary storage means in advance. By collating with the information, it is determined whether or not a predetermined relationship is established between the two information. For example, whether or not the input face image is similar to the dictionary information is evaluated. If the similarity is high, the person to be verified is determined to be a registrant (person), and if the similarity is low Judge as unregistered (not the person).
JP-A-9-251534 Japanese Patent Application Laid-Open No. 64-79895

However, since the conventional face collation apparatus performs face collation using only the face image obtained from the camera as described above, for example, it is reliably excluded that the person is impersonated by a person's face photograph or video image. It is impossible.
Further, since face matching is performed only by image processing, for example, there is a problem that twin face discrimination is impossible.

  Therefore, the present invention can reliably eliminate face impersonation by a face photograph or a video image, and can also identify a face that cannot be identified by face matching only by image processing, and a traffic An object is to provide a control device.

  The face collation apparatus according to the present invention includes an image acquisition unit that acquires a face image of a person to be verified as a verification target, a first dictionary information based on a face image acquired by the image acquisition unit and a face image registered in advance. And a face image acquired by the image acquisition unit as a result of the verification by the first verification unit, and a first verification unit that determines whether or not a predetermined relationship is established between both pieces of information. Temperature distribution measuring means for obtaining temperature distribution information by measuring the temperature distribution of at least the face portion of the person to be verified, and the temperature distribution measuring means when a predetermined relationship is established between the first dictionary information and the first dictionary information. A second determination is made as to whether or not a predetermined relationship is established between the two pieces of information by collating the obtained temperature distribution information with the second dictionary information based on the temperature distribution of the human face part registered in advance. With verification means To have.

  Further, the traffic control device of the present invention includes image acquisition means for acquiring a face image of a passerby to be collated, and first dictionary information based on a face image acquired by the image acquisition means and a face image registered in advance. And a face obtained by the image obtaining means as a result of the collation by the first collating means and a first collating means for judging whether or not a predetermined relationship is established between both pieces of information. When a predetermined relationship is established between the image and the first dictionary information, temperature distribution measuring means for measuring temperature distribution of at least the face portion of the passerby to obtain temperature distribution information, and this temperature distribution measuring means A second determination is made as to whether or not a predetermined relationship is established between the two pieces of information by collating the obtained temperature distribution information with the second dictionary information based on the temperature distribution of the human face part registered in advance. Matching means; It is provided with a passage control means for controlling the passage of the passer based on the result of the comparison by the second comparing means for.

  According to the present invention, face collation based on the temperature distribution of the face is also performed, so that it is possible to surely eliminate the identity impersonation by a face photograph or a video image, and the face collation only by image processing cannot be discriminated. It is possible to provide a face collation device and a traffic control device that can discriminate between various faces.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 schematically shows a configuration of an entrance / exit management system to which the face matching device and the traffic control device according to the present embodiment are applied. This entrance / exit management system includes a camera 101 serving as an image acquisition unit that captures and inputs a face image (an image including at least a face) of a person to be verified (hereinafter also referred to as a person) 100, and at least the person 100 to be verified. The thermography 102 as a temperature distribution measuring means for measuring the temperature distribution of the face portion to obtain temperature distribution information, the face image input to the person to be verified 100, various guidance, etc. are displayed, or the person 100 to be verified is specified A facial feature as a facial feature acquisition unit that inputs an input number information (face feature information) unique to the person by processing an image input from the input / output unit 103 and the camera 101 that inputs an ID number, a password, and the like. The acquisition unit 104 stores a first dictionary description that stores (registers) registered feature information (face feature information) based on a face image of a person registered in advance as first dictionary information. The unit 105 stores (registers) the temperature distribution pattern in the human face portion as second dictionary information, and stores (registers) the temperature distribution pattern in the face portion of the person to be registered that has been registered in advance as third dictionary information. ) By comparing the input feature information obtained from the second dictionary storage unit 106 and the face feature acquisition unit 104 with the first dictionary information in the first dictionary storage unit 105, a predetermined amount between the two information It is determined whether or not the relationship is established, and the temperature distribution information of the face portion obtained from the thermography 102 is compared with the second dictionary information or the third dictionary information in the second dictionary storage unit 106. The collation unit 107 as collation means for determining whether or not a predetermined relationship is established between the two information, and the entrance / exit door 109 of the important facility 108 according to the determination result of the collation unit 107 Closing control door control section 110 as passage control means for, and is configured to have the input and output control unit 111 for controlling input and output unit 103.

The camera 101 captures and inputs a face image of the person to be verified 100 (an image including at least a face), and includes, for example, a television camera using an image sensor such as a CCD sensor.
The thermography 102 measures, for example, the temperature distribution of at least the face portion of the person to be verified 100, particularly the front face portion, using infrared rays or the like.

  The input / output unit 103 displays the face image of the person to be verified 100 captured by the camera 101, various guidance, processing results, and the like, and inputs the ID number and password of the person to be verified 100. The liquid crystal display unit and the touch panel are integrally configured, and are provided in the vicinity of the camera 101, for example, in the upper part of the camera 101, at a position substantially corresponding to the face portion of the person to be verified 100. Note that not only the guidance by display but also a voice guidance function may be added.

  The face feature acquisition unit 104 detects a person's face in the image input from the camera 101 and extracts (acquires) input feature information. As a face detection method, for example, literature (Kazuhiro Fukui, Osamu Yamaguchi: “Face feature point extraction by combination of shape extraction and pattern matching”, IEICE Transactions (D), vol. J80-D-II, No. 8, pp 2170-2177 (1997)) can be used to realize highly accurate face detection. In this case, a location having the highest correlation value is determined as a face region by obtaining a correlation value while moving a template prepared in advance in the image.

  The first dictionary storage unit 105 stores, as first dictionary information, face feature information of a person who is permitted to use in the entire system in advance. As the first dictionary information to be registered, the input face image, extracted feature amount, subspace, eigenvector for configuring the subspace, correlation matrix, status information such as time, date and time at the time of registration, Personal information such as a personal identification number and ID number can be listed, and can be stored as necessary. In the present embodiment, the face cut-out image acquired by the face feature acquisition unit 104 is used. Vector information obtained by converting to eigenspace by KL expansion is stored, and this information is handled as dictionary information.

  The second dictionary storage unit 106 stores an average temperature distribution pattern in the human face portion as second dictionary information, and stores the temperature distribution pattern in the face portion of a person permitted to be used in the entire system in advance. 3 is stored as dictionary information.

  In this case, the second dictionary information and the third dictionary information are at least one of environmental conditions such as season, time, temperature, etc., in this example, for example, every season (spring, summer, autumn, winter) The second dictionary information and the third dictionary information corresponding to the season in which the face matching device is operated are selected and used for matching.

  The third dictionary information is registered, for example, by measuring with the thermography 102 when the first dictionary information is registered. However, when the first dictionary information is registered with a facial photograph. Alternatively, it may be registered at the time of the first face collation.

  The collation unit 107 collates the input feature information obtained from the face feature acquisition unit 104 with the first dictionary information in the first dictionary storage unit 105, whereby a predetermined relationship is established between the two pieces of information. For example, the similarity between the input feature information and the first dictionary information is obtained, and when the obtained similarity is higher than a threshold, the person is determined to be a registrant (person) and obtained. If the similarity is lower than the threshold, it is determined that the person is not registered (not the person). As this determination method, one-to-one verification of whether or not the first dictionary information specified by using the ID card or numeric keypad input by the input / output unit 103 matches, and all or a part of registrants. It is possible to use both 1-to-N collation in which a match determination is made for the brute force and if there is at least one matching first dictionary information, it is determined as a registrant.

  In addition, the collation unit 107 uses the temperature distribution information of the face portion obtained from the thermography 102 corresponding to the face area detected by the face feature acquisition unit 104 and the second dictionary information in the second dictionary storage unit 106. By collating, whether or not a predetermined relationship is established between both pieces of information, for example, the degree of similarity between the temperature distribution information and the second dictionary information is obtained, and when the obtained degree of similarity is higher than a threshold value If it is determined that the face is a human face and the obtained similarity is lower than the threshold, it is determined that the face is not a human face.

  Furthermore, the collation unit 107 uses the temperature distribution information of the face portion obtained from the thermography 102 corresponding to the face area detected by the face feature acquisition unit 104 and the third dictionary information in the second dictionary storage unit 106. By collating, whether or not a predetermined relationship is established between both pieces of information, for example, the degree of similarity between the temperature distribution information and the third dictionary information is obtained, and when the obtained degree of similarity is higher than a threshold value It is determined that the person is a registered person (person), and if the obtained similarity is lower than the threshold value, the person is determined as an unregistered person (not the person).

  And the collation part 107 outputs the signal of "door opening" to the door control part 110, for example, when the calculated similarity degree is more than a threshold value, and when the calculated similarity degree is smaller than a threshold value, the door control part 110 Outputs a “door closed” signal. Upon receiving the “door open” signal from the verification unit 107, the door control unit 110 controls the entrance / exit door 109 to be in an open state, permits entry of the verified person (passerby) 100, and displays “door closed”. ”Is held, the entrance / exit door 109 is kept closed, and entry of the person to be verified 100 is rejected.

Next, the operation of such a configuration will be described with reference to the flowchart shown in FIG.
The person to be verified 100 stands in front of this apparatus (camera 101), and by visually observing the screen of the input / output unit 103, the camera 101 captures and inputs an image of the person 100 to be verified (step S1). When the face image of the person to be verified 100 is input by the camera 101, the verification unit 107 includes the input feature information acquired by the facial feature acquisition unit 104 and the first dictionary stored in the first dictionary storage unit 105. By collating the information, the similarity between the input feature information and the first dictionary information is obtained, and when the obtained similarity is higher than the threshold, the person is determined to be a registrant (person). If the obtained similarity is lower than the threshold value, it is determined that the person is not a registrant (person) (step S2). Here, the process of step S2 corresponds to the first collating means in the present invention.

  If it is determined in step S2 that the user is not a registrant, the result is collation NG, a message to that effect is displayed on the screen of the input / output unit 103, an alarm is output to a monitoring panel (not shown), and the door control unit 110 is further displayed. A “door closed” signal is output (step S3).

  If it is determined in step S2 that the user is a registrant, the collation unit 107 determines the current season by referring to, for example, a calendar function (not shown), and stores the second dictionary information corresponding to the determined season. 2 is selected from the dictionary storage unit 106, and the selected second dictionary information is collated with the temperature distribution information of the face portion obtained from the thermography 102 corresponding to the face area detected by the face feature acquisition unit 104. Thus, the similarity between the temperature distribution information and the second dictionary information is obtained, and when the obtained similarity is higher than the threshold, it is determined to be a human face, and the obtained similarity is lower than the threshold. In this case, it is determined that the face is not a human face (step S4). Here, the process of step S4 corresponds to the second collating means in the present invention.

  If it is determined in step S4 that the face is not a human face, the result is NG, a message to that effect is displayed on the screen of the input / output unit 103, an alarm is output to a monitoring panel (not shown), and the door control unit 110 is further displayed. On the other hand, a “door closed” signal is output (step S3).

  If it is determined in step S4 that the face is a human face, the collation unit 107 determines whether or not to perform face collation based on the temperature distribution (step S5). That is, a setting switch (not shown) for setting whether or not to perform face collation based on the temperature distribution is provided, and determination is made by referring to the setting contents of this setting switch.

  If it is determined in step S5 that face collation based on the temperature distribution is not performed, collation is OK, a message to that effect is displayed on the screen of the input / output unit 103, and a “door open” signal is output to the door control unit 110. (Step S6).

  If it is determined in step S5 that face matching based on the temperature distribution is to be performed, the matching unit 107 determines the current season in the same manner as in step S2, and sets the third dictionary information corresponding to the determined season to the second dictionary. By selecting from the storage unit 106 and collating the selected third dictionary information with the temperature distribution information of the face portion obtained from the thermography 102 corresponding to the face area detected by the face feature acquisition unit 104 The similarity between the temperature distribution information and the third dictionary information is obtained. If the obtained similarity is higher than the threshold, the person is determined to be a registrant (person), and the obtained similarity is greater than the threshold. If it is lower, it is determined that the user is not a registered person (step S7). Here, the process of step S7 corresponds to the third matching means in the present invention.

  If it is determined in step S7 that the user is not a registrant, the result is collation NG, the fact is displayed on the screen of the input / output unit 103, an alarm is output to a monitoring panel (not shown), and the door control unit 110 is further displayed. A “door closed” signal is output (step S3).

  If it is determined in step S7 that the user is a registrant, collation is OK, a message to that effect is displayed on the screen of the input / output unit 103, and a “door open” signal is output to the door control unit 110 (step S6). .

  In this way, by authenticating the person by combining the face collation based on the face image of the person being collated and the face collation based on the human face temperature distribution, it is possible to surely eliminate identity impersonation due to a face photo or video image. Can do. In addition, it is possible to discriminate faces (for example, twins) that cannot be discriminated by face collation using only image processing.

1 is a block diagram schematically showing a configuration of an entrance / exit management system to which a face matching device and a traffic control device according to an embodiment of the present invention are applied. The flowchart explaining operation | movement.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 100 ... Person to be verified (person) 101 ... Camera (image acquisition means) 102 ... Thermography (temperature distribution measurement means) 103 ... Input / output unit 104 ... Face feature acquisition unit 105 ... First dictionary storage unit 106: second dictionary storage unit, 107: collation unit (collation unit), 110: door control unit (traffic control unit).

Claims (6)

Image acquisition means for acquiring a face image of a person to be verified to be verified;
First collation for determining whether or not a predetermined relationship is established between both pieces of information by collating the face image acquired by the image acquisition means with the first dictionary information based on the face image registered in advance. Means,
When a predetermined relationship is established between the face image acquired by the image acquisition unit and the first dictionary information as a result of the verification by the first verification unit, the temperature distribution of at least the face portion of the person to be verified Temperature distribution measuring means for obtaining temperature distribution information by measuring
Whether or not a predetermined relationship is established between the two information by collating the temperature distribution information obtained by the temperature distribution measuring means with the second dictionary information based on the temperature distribution of the human face part registered in advance. Second verification means for determining
A face collating apparatus comprising:   The second dictionary information is registered for each of at least one of environmental conditions such as season, time, temperature, etc., and second dictionary information corresponding to the environmental conditions in which the face matching device is operated. The face matching apparatus according to claim 1, wherein is selected and used for matching.   When the predetermined relationship is established between the temperature distribution information obtained by the temperature distribution measuring means and the second dictionary information as a result of the collation by the second collating means, the temperature distribution measuring means obtained Third collation for determining whether or not a predetermined relationship is established between both pieces of information by collating the temperature distribution information with the third dictionary information based on the temperature distribution of the face portion of the person to be collated registered in advance. The face collating apparatus according to claim 1, further comprising means.   The third dictionary information is registered for each of at least one environmental condition such as season, time, temperature, etc., and the third dictionary information corresponding to the environmental condition in which the face matching device is operated. 4. The face matching apparatus according to claim 3, wherein is selected and used for matching. Image acquisition means for acquiring a face image of a passerby to be verified;
First collation for determining whether or not a predetermined relationship is established between both pieces of information by collating the face image acquired by the image acquisition means with the first dictionary information based on the face image registered in advance. Means,
When a predetermined relationship is established between the face image acquired by the image acquisition unit and the first dictionary information as a result of the verification by the first verification unit, the temperature distribution of at least the face portion of the passerby is determined. Temperature distribution measuring means for obtaining temperature distribution information by measuring;
Whether or not a predetermined relationship is established between the two information by collating the temperature distribution information obtained by the temperature distribution measuring means with the second dictionary information based on the temperature distribution of the human face part registered in advance. Second verification means for determining
A passage control means for controlling the passage of the passer based on the result of the collation by the second collating means;
A traffic control device comprising: When the predetermined relationship is established between the temperature distribution information obtained by the temperature distribution measuring means and the second dictionary information as a result of the collation by the second collating means, the temperature distribution measuring means obtained Third collation for determining whether or not a predetermined relationship is established between the two information by collating the temperature distribution information with the third dictionary information based on the temperature distribution of the face portion of the person to be collated registered in advance. Further comprising means,
The traffic control device according to claim 5, wherein the traffic control unit controls the traffic of the passer based on a result of verification by the third verification unit.

Images