JP2005135271A - Face collation device and passage controller - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a face collation device and a passage controller capable of determining whether a face image taken by a camera part 101 is a spoofed image using a photograph and the like or not for preventing a masquerade by another person by means of a simple structure. <P>SOLUTION: A distance between the camera part 101 and a collation objective person is measured by a distance measuring sensor part 105, and based on the distance between the camera part 101 and the collation objective person measured by the distance measuring sensor 105 and a distance between a plurality of characteristic points on a face in an image captured by the camera part 101, the size of the face photographed by the camera part 101 is estimated. It is determined whether the size of the face is appropriate or not, and it is determined that the face size is not appropriate, collation processing based on the image is not allowed. Only when the face size is determined to be appropriate, collation processing based on the face image is carried out. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

  The present invention includes, for example, a face collation device and a traffic control device that perform collation processing between an image (face image) including at least a face of a person to be collated photographed by a camera or the like and a face image registered in a storage device in advance. .

  2. Description of the Related Art Conventionally, there is a face matching device that performs a matching process between an image (face image) including at least a face of a person to be verified photographed by a camera and a face image previously registered in a storage device. In such a face collation apparatus, it is necessary to prevent impersonation of another person using a photograph or the like in order to realize highly accurate collation processing. For example, a face collation apparatus disclosed in Japanese Patent Application Laid-Open No. 2001-338296 includes a first illumination unit that irradiates a face of a person to be collated photographed by a camera from the upper right or upper left of the camera, and the camera. And a second illumination unit that illuminates the face of the person to be verified to be photographed from below the camera. That is, in the face collation apparatus disclosed in Japanese Patent Laid-Open No. 2001-338296, the face of the person to be collated illuminated by the first and second illumination units is photographed by a camera, and the face image photographed by the camera The feature amount of the face of the person to be collated is extracted from this, and the extracted feature amount is collated with the reference feature amount for the face image registered in advance, thereby preventing spoofing of others.

However, in the conventional face collation apparatus having a plurality of illuminations as described above, the entire apparatus becomes complicated and the apparatus itself becomes expensive, making it difficult to provide an inexpensive apparatus, or lighting a plurality of illuminations. Since the photographed face image and the registered face image are collated, there is a problem that the process becomes complicated and the collation process takes time.
JP 2001-338296 A

  As described above, the conventional face matching device has a problem that the process is complicated and time-consuming in order to prevent impersonation by another person using a photograph or the like. Therefore, an object of the present invention is to provide a face matching device and a traffic control device that can prevent impersonation of another person with high accuracy with a simple device configuration without taking time for the matching processing. .

  The face collation apparatus according to the present invention captures a face image of a person to be collated in a storage unit in which a registrant's face image is registered in advance and a face image registered in the storage unit. A photographing means for detecting, a detecting means for detecting a distance from the photographing means to a person to be collated for photographing a face image by the photographing means, a distance from the photographing means to the person to be collated detected by the detecting means, and the photographing Based on the distance between feature points of a plurality of faces in the face image photographed by the means, a judging means for judging the size of the face photographed by the photographing means, and photographing by the photographing means judged by the judging means Determining means for determining whether or not the face size in the captured face image is appropriate, and determining by the determining means that the face size in the face image photographed by the photographing means is appropriate In the case of the face image captured by the image capturing unit and the execution unit that performs a collation process between the face image of the registrant registered in the storage unit and the face image captured by the image capturing unit by the determination unit When it is determined that the size of the face is not appropriate, the image processing apparatus includes processing means for disabling the collation process between the face image photographed by the photographing means and the registrant face image registered in the storage means.

  The face collation apparatus according to the present invention captures a face image of a person to be collated in a storage unit in which a registrant's face image is registered in advance and a face image registered in the storage unit. A photographing means for detecting, a detecting means for detecting a distance from the photographing means to a person to be collated for photographing a face image by the photographing means, a distance from the photographing means to the person to be collated detected by the detecting means, and the photographing Based on the distance between feature points of a plurality of faces in the face image photographed by the means, a judging means for judging the size of the face photographed by the photographing means, and photographing by the photographing means judged by the judging means First determination means for determining whether or not the face size in the face image is appropriate, and the face size in the face image photographed by the photographing means by the first determination means is appropriate When it is determined that there is an illumination unit that sequentially irradiates light from the left side and the right side of the imaging unit to the verification target person, and when the light is irradiated from the left side of the imaging unit by the illumination unit, A second determination means for determining whether or not the shadow appearing on the face and the shadow appearing on the face of the person to be collated when the illumination means irradiates light from the right side of the photographing means; When it is determined by the determining means that the shadow appearing on the face of the person to be verified is normal, a matching process between the face image photographed by the photographing means and the face image of the registrant registered in the storage means is executed. When the face size of the face image photographed by the photographing means is not appropriate by the executing means and the first judging means, or appears on the face of the person to be collated by the second judging means Normal shadow If not, and a processing means to disable the verification process of the registrant's face image registered in the storage means and photographed face image by the imaging means.

  The traffic control device according to the present invention is based on a result of collation between a storage means in which a face image of a person permitted to pass in advance is registered and a face image of a passer-by and a face image registered in the storage means. And a detection means for detecting a distance from the imaging means to a passerby who took a face image by the imaging means, and this detection Based on the distance from the photographing means to the passerby detected by the means and the distance between the feature points of a plurality of faces in the face image photographed by the photographing means, the face of the face image photographed by the photographing means A judging means for judging the size, a judging means for judging whether or not the face size in the face image photographed by the photographing means judged by the judging means is appropriate, and the judging means If it is determined that the face size in the face image photographed by the photographing means is appropriate, the face image photographed by the photographing means and the face of a person permitted to pass registered in the storage means When the matching between the execution unit that executes the matching process with the image and the face image captured by the shooting unit executed by the execution unit and the face image of the person registered in the storage unit is successful, A pass permission means for allowing a passer to pass, and a face image photographed by the photographing means executed by the execution means and a face image of a person registered in the storage means fail, or And a passage blocking unit that blocks the passer-by when the determination unit determines that the size of the face in the face image captured by the shooting unit is not appropriate.

  The traffic control device according to the present invention is based on a result of collation between a storage means in which a face image of a person permitted to pass in advance is registered and a face image of a passer-by and a face image registered in the storage means. And a detection means for detecting a distance from the imaging means to a passerby who took a face image by the imaging means, and this detection The size of the face in the face image photographed by the photographing means based on the distance from the photographing means detected by the means to the passerby and the distance between the feature points of a plurality of faces in the face image photographed by the photographing means A first judging means for judging whether or not the size of the face in the face image photographed by the photographing means judged by the judging means is appropriate; An illumination unit that sequentially irradiates light from the left side and the right side of the imaging unit with respect to the person to be collated, A shadow that appears on the face of the person to be collated when light is emitted from the left side of the photographing means by the illumination means and a shadow that appears on the face of the person to be collated when light is irradiated from the right side of the photographing means by the illumination means. A second determination unit that determines whether or not the image is normal, and a face image captured by the imaging unit when the second determination unit determines that the shadow appearing on the face of the person to be collated is normal Execution means for executing a collation process with a registrant's face image registered in the storage means, a face image photographed by the photographing means executed by the execution means, and a person registered in the storage means of When the comparison with the image is successful, a passage permission means for permitting passage of the passer-by, a face image photographed by the photographing means executed by the execution means, and a person's face registered in the storage means When collation with the image fails, when the first determination unit determines that the face size in the face image captured by the imaging unit is not appropriate, or when the second determination unit determines the passer's When it is determined that the shadow appearing on the face is not normal, a passage blocking means for blocking the passage of the passer-by is provided.

  According to the present invention, it is possible to provide a face matching device and a traffic control device that can prevent impersonation of another person with high accuracy with a simple device configuration without taking time for the matching processing.

  The best mode for carrying out the present invention will be described below in detail with reference to the drawings.

First, the first embodiment will be described.
FIG. 1 is a diagram showing an external configuration of a face collation apparatus according to the first embodiment of the present invention.

  As shown in FIG. 1, the face collation apparatus 1 includes a camera unit 101, a control unit 102, a display unit 103, an operation unit 104, a distance measurement sensor unit 105, a speaker unit 106, an illumination unit 107, a signal control unit 108, and the like. have. Further, the face matching device 1 is connected to a door opening / closing mechanism 109a that opens and closes the door 109 to constitute a traffic control device.

  The camera unit 101 is configured by a digital video camera or the like, and captures still images, moving images, or continuous images. This camera unit 101 is provided below the display unit 103, and an image (hereinafter referred to as a face image) of an area including at least the face of a user (person to be verified) who is viewing the display unit 103 is slightly faced. Shooting from the lower right side of the front. The display unit 103 is configured by a liquid crystal display device or the like, and is installed at the approximate center of the face matching device 1. The display unit 103 displays an image being shot by the camera unit 101 or displays guidance for a user. Further, the speaker unit 106 outputs guidance for the user and the like by voice.

  The operation unit 104 includes a numeric keypad. The operation unit 104 is for inputting, for example, an ID number or password as identification information given to each user in order for the user to identify an individual.

  The illumination unit 107 illuminates the imaging range of the camera unit 101 when the camera unit 101 captures a user's face image. The illumination unit 107 is turned on when, for example, the surrounding illuminance is insufficient.

  The distance measuring sensor unit 105 is installed in front of the face matching device 1 and measures the distance between the face matching device 1 and the person to be checked. The distance measuring sensor unit 105 measures, for example, the distance between the face matching device 1 and the person to be verified in the horizontal direction. The distance measuring sensor unit 105 is a distance measuring sensor using ultrasonic waves or the like. In the distance measuring sensor unit 105 using ultrasonic waves, for example, the ultrasonic wave is transmitted toward an object (a person to be verified), and the reflection is received, thereby using the Doppler effect, and the distance measuring sensor unit 105 The distance from 105 to the object is detected.

  The control unit 102 controls the overall operation of the face collation apparatus 1. The control unit 102 includes a storage device 102a in which an ID number, a password, a face image, a face size, a face feature amount, and the like regarding a registrant are registered in advance as dictionary data.

  The control unit 102 also includes a processor unit (not shown) that performs arithmetic processing, a program memory unit (not shown) that stores a control program executed by the processor unit, and temporarily stores work data. A work memory unit (not shown), an image memory unit (not shown) for mainly recording image data, various drive control units (not shown) for controlling the units, and data captured by the camera unit 101. A capture unit (not shown) that captures image information is included.

  The signal control unit 108 is provided when the face matching device 1 is used as a traffic control device. That is, the signal control unit 108 controls an opening / closing signal to the door opening / closing mechanism unit 109 a provided in the door 109 based on the matching result by the face matching device 1. The door opening / closing mechanism 109a controls the opening / closing of the door 109 based on a control signal supplied from the signal controller 108.

  Next, the operation of the face matching device 1 configured as described above will be described.

  FIG. 2 is a flowchart for explaining the operation of the face collating apparatus 1.

  Here, it is assumed that the distance measuring sensor unit 105 measures the distance to the object existing in front of the face matching device 1 (step S11). In this state, when the person to be collated appears in front of the face collation apparatus 1, the value measured by the distance measurement sensor unit 105 varies (YES in step S12).

  For example, when there is a wall in front of the face matching device 1, the distance measuring sensor unit 105 measures the distance to the wall existing in front of the face matching device 1 in a state where there is no person to be checked. doing. When a verification target person appears in front of the face verification device 1, the distance measuring sensor unit 105 measures the distance to the verification target person. As described above, in the face collation apparatus 1, it is determined whether or not a person to be collated has appeared in front of the face collation apparatus 1 based on the distance measured by the distance measuring sensor unit 105.

  When the distance measured by the distance measurement sensor unit 105 varies (step S12, YES), the distance measurement sensor unit 105 notifies the control unit 102 that a verification target person has appeared. Upon receiving this notification, the control unit 102 starts capturing an image (or capturing image data) with the camera unit 101 (step S13). An image captured by the camera unit 101 is captured as image data by a capture unit (not shown) in the control unit 102.

  When the operation unit 104 is configured to allow the verification target person to input an ID number, a password, or the like, when the verification target person inputs the ID number or password, the camera unit It is also possible to start taking an image using the image 101 and capture it as image data in the control unit 102.

  When the image captured by the camera unit 101 is captured, the processor (not shown) of the control unit 102 performs face detection processing for detecting a human face region (face region) from the captured image data (step S14). ). In this face detection process, for example, it is determined whether a face has been detected based on whether an image of both eyes and a nose image have been detected from the image captured from the camera unit 101.

  When a face cannot be detected from the image captured from the camera unit 101 in the face detection process (step S15, NO), the processor of the control unit 102 returns to step S11 and repeats the same process. Further, when the face cannot be detected by the face detection process, the processor of the control unit 102 may fetch the image data from the camera unit 101 again and execute the face detection process again.

  When a face can be detected from the image captured from the camera unit 101 by the face detection process (step S15, YES), the processor unit of the control unit 102 extracts a facial feature point from the detected face region. Is performed (step S16). Here, both eyes and nostril are detected as facial feature points.

  In the facial feature point extraction process, a pupil region and a nostril region that can be regarded as a substantially circular region are detected in the face region detected in step S14, and the center positions of these detected regions are used as facial feature points. Extract. In addition, as a process of the facial feature points, other parts of the face such as eyebrows may be detected as the feature points in addition to both eyes and nostrils.

  When facial feature points are extracted by the facial feature point extraction process (step S16), the processor of the control unit 102 performs a process of calculating a distance X from the camera unit 101 to the facial feature points (step S16). S17).

  As shown in FIG. 3, the face X in the horizontal direction (direction measured by the distance measuring sensor) is used as the distance X to the face feature point located at the approximate center on the image photographed by the camera unit 101. 1 is an angle formed by the optical axis of the camera unit 101 with respect to the horizontal direction (direction measured by the distance measuring sensor) and the distance l from the distance measurement sensor unit 105 to the verification target person as a distance from 1 to the verification target person. Is calculated by the following equation (1) using the elevation angle θ of the camera.

X = lcos θ (1)
The distance from the camera unit 101 to each feature point of the face is the distance l measured by the distance measuring sensor, the straight line connecting the camera unit 101 and each feature point of the face, and the horizontal direction. It is calculated based on the angle. For this reason, in the present embodiment, the distance to the feature point (for example, nostril) of the face located at the approximate center on the image photographed by the camera unit 101 is the camera unit 101 on the optical axis of the camera unit 101. The distance from the feature point of the face to the face (optical axis distance) X.

  When the distance X between the camera unit 101 and the facial feature point is calculated by the above equation (1), the processor of the control unit 102 calculates the face size Y in the image photographed by the camera unit 101. (Step S18).

  The face size Y is calculated based on the distance X from the camera unit 101 calculated in step S17 to the feature point of the face and the distance between each feature point of the face in the face image. For example, assuming that the distance X from the camera unit 101 to the nostril as a facial feature point is calculated in step S17, the processor unit of the control unit performs facial feature points other than the nostrils (for example, both eyes and eyebrows). Etc.) is calculated from the image data acquired from the camera unit 101.

  That is, by the distance from the camera unit 101 to the face of the person to be collated photographed by the camera unit 101 and the distance between feature points of a plurality of faces in the image photographed by the camera unit 101, the camera unit The actual size of the face photographed at 101 is calculated. By such processing, the actual face size of the person to be collated photographed by the camera unit 101 can be inferred, and it can be determined whether or not the face size is appropriate.

  That is, when calculating the actual face size Y of the person to be collated photographed by the camera unit 101 in step S18, the processor of the control unit 102 determines whether or not the calculated face size Y is appropriate. Is determined (step S19). In this determination, the calculated face size Y is compared with the face size of the registrant registered in advance in the storage device 102a in the control unit 102. For example, the calculated face size Y is stored in the storage device. It is determined whether or not the calculated face size is appropriate depending on whether or not the face size registered in 102a is within a predetermined allowable range.

  When dictionary data of a plurality of registrants is registered in the storage device 102a, it is determined whether or not the face size Y calculated for all registrants (dictionary data) is appropriate in the determination. It may be determined, or the face size is appropriate by comparing the face size of the registrant who has been successfully matched after executing the face matching process with the calculated face size Y. It may be determined whether or not.

  If the ID number or password from the person to be collated has been input first, the registrant is narrowed down by the ID number or password entered by the person to be collated, and the size of the registrant's face is determined. The calculated face size Y may be compared to determine whether the face size is appropriate.

  When it is determined by the above determination that the face size Y is not appropriate (step S19, NO), the processor of the control unit 102 confirms the validity of the face image captured by the camera unit 101 by the display unit 103. Since it is not possible, a message indicating that collation is impossible is displayed (step S20), and the signal control unit 108 outputs a door closing control signal for keeping the door closed to the door opening / closing mechanism 109a. Receiving this door closing control signal, the door opening / closing mechanism 109a keeps the door 109 closed and blocks the person to be verified from passing (step S21).

  Thereby, when the face size in the image photographed by the camera unit 101 is significantly different from the registered face size, there is a possibility that the person to be collated is “spoofing” another person using a photograph or the like. It can be determined that the person is present, and collation processing for the person can be disabled, or the door can be closed to prevent the person from passing.

  If it is determined by the above determination that the face size Y is appropriate, the processor of the control unit 102 causes the face image of the registrant registered in the storage device 102 a and the face image captured by the camera unit 101. (Step S22).

  In the collation process, when dictionary data of a plurality of registrants are registered in the storage device 102a, the similarity with all registrants (dictionary data) is determined for the face image photographed by the camera unit 101. If the degree of similarity with the most similar registrant is greater than or equal to a predetermined value, it is determined that the collation with the registrant is successful. Further, when the verification target person inputs an ID number or password, the registrant is narrowed down by the ID number or password input by the verification target person, and the registrant's face image and the camera unit 101 are used. The degree of similarity with the photographed face image is calculated, and if the degree of similarity is equal to or greater than a predetermined value, it is determined that the collation with the registrant is successful.

  When the collation between the face image photographed by the camera unit 101 and the registrant's face image has succeeded by the collation process (step S23, YES), the processor of the control unit 102 succeeds in collation by the display unit 103. In addition to displaying a guidance to the effect (step S24), the signal control unit 108 outputs a door opening control signal for opening the door to the door opening / closing mechanism unit 109a (step S25). Upon receiving this door opening control signal, the door opening / closing mechanism 109a opens the door 109 and allows the person to be verified to pass.

  When the collation between the face image captured by the camera unit 101 and the registered face image fails due to the collation process (step S3, NO), the processor of the control unit 102 performs the collation by the display unit 103. In addition to displaying the failure (step S26), the signal control unit 108 outputs a door closing control signal for keeping the door closed to the door opening / closing mechanism 109a (step S21). Upon receiving this door closing control signal, the door opening / closing mechanism 109a keeps the door 109 closed and prevents the person to be verified from passing.

  As described above, according to the first embodiment, the distance between the camera and the verification target person is measured by the distance measurement sensor, the distance between the camera and the verification target person measured by the distance measurement sensor, and the above Analogize the size of the face photographed with the camera based on the distance between the feature points of a plurality of faces in the image photographed with the camera, determine whether the face size is appropriate, If it is determined by this determination that the face size is not appropriate, the matching process using the image is disabled, and only when the face size is determined to be appropriate by the determination, the matching process using the face image is performed. It is a thing.

  As a result, it is possible to determine whether the face image captured by the camera is impersonation using a photograph or the like, and it is possible to provide a face matching device that can prevent impersonation by another person with a simple configuration. .

  Next, a second embodiment will be described.

  FIG. 4 is a diagram showing an external configuration of the face matching device 2 according to the second embodiment of the present invention.

  As shown in FIG. 4, the face matching device 2 includes a camera unit 101, a control unit 102, a display unit 103, an operation unit 104, a distance measurement sensor unit 105, a speaker unit 106, a first illumination unit 107, and signal control. Part 108, the second illumination part 201 (201a, 201b), and the like. Further, the face matching device 1 is connected to a door opening / closing mechanism 109a that opens and closes the door 109 to constitute a traffic control device.

  The face matching device 2 according to the second embodiment shown in FIG. 4 has a configuration in which the second illumination units 201a and 201b are added to the face matching device 1 according to the first embodiment shown in FIG. Yes. For this reason, about the structure similar to the said face collation apparatus 1, the same code | symbol is attached | subjected to the same location and detailed description is abbreviate | omitted.

  The lighting of the second illumination units 201 a and 201 b is controlled by the processor of the control unit 102. The second illumination unit (left illumination unit) 201a is installed on the left side of the camera unit 101 and irradiates light from the left side of the camera unit 101 to the face of the person to be collated photographed by the camera unit 101. Is. The second illumination unit (right illumination unit) 201b is installed on the right side of the camera unit 101, and from the right side of the camera unit 101 with respect to the face of the person to be collated photographed by the camera unit 101. It irradiates light.

  Next, the operation of the face matching device 2 configured as described above will be described in detail.

  FIG. 5 is a flowchart for explaining the operation of the face matching device 2.

  The operation of the face collation device 2 according to the second embodiment shown in FIG. 5 is the same as the operation of the face collation device 1 according to the first embodiment shown in FIG. In this case, the operation (steps S31 to S33) for determining whether or not the shadow formed on the face of the person to be collated is normal when the (part) is turned on is added. For this reason, in the description of the operation of the face collation apparatus 2 shown in FIG. 5, the same operation as the operation of the face collation apparatus 1 is given the same step and the detailed description is omitted.

  That is, in the face matching device 2, first, the operations of steps S11 to S19 described in the first embodiment are executed. When it is determined in step S19 that the face size is appropriate, the processor of the control unit 102 lights the left illumination unit 201a as the second illumination unit installed on the left side of the camera unit 101. Then, the face image of the person to be collated is photographed by the camera unit 101 (step S31).

  In this case, the face of the person to be collated photographed by the camera unit 101 is irradiated with light from the left side of the face collation apparatus 2 (the right side of the person to be collated). For this reason, in the camera unit 101, if the face of the person to be collated is in a normal state (that is, not “spoofing” by a photograph or the like), a shadow is formed on the left side of the face of the person to be collated. A face image is taken.

  When the capture unit captures a face image taken with the left illumination unit 201a turned on, the processor of the control unit 102 next turns on the right illumination unit 201b as the second illumination unit, The camera unit 101 captures the face image of the person to be verified (step S32).

  In this case, the face of the person to be collated photographed by the camera unit 101 is irradiated with light from the right side (left side of the person to be collated) of the face collation apparatus 2. For this reason, in the camera unit 101, if the face of the person to be collated is in a normal state (that is, not “spoofing” by a photograph or the like), a shadow is formed on the right side of the face of the person to be collated. A face image is taken.

  When a capture unit captures a face image captured with the right illumination unit 201b turned on, the processor of the control unit 102 turns on the left illumination unit 201a and shadows appearing on the captured face image It is determined whether or not the shadow appearing in the face image photographed by turning on the right illumination unit 201b is in a normal state (step S33).

  In this determination, a shadow appears on the left side of the face of the person to be collated in the face image taken with the left illumination unit 201a turned on, and the object to be collated in the face image taken with the right illumination unit 201b turned on Whether or not a shadow is formed in the face image is determined based on whether or not a shadow appears on the right side of the person's face.

  If it is determined by this determination that the shadow is normal (step S33, YES), the processor of the control unit 102 proceeds to step S22 and executes collation processing. If it is determined by the above determination that the shadow is not normal (step S33, NO), the processor of the control unit 102 proceeds to step S20, displays that the verification is impossible, and disables the door opening. .

  Next, the determination process in step S33 will be described with reference to a specific example.

  6 (a) and 6 (b) show examples of face images when the shadows are normal, and FIGS. 7 (a) and 7 (b) show that the shadows are not normal. The example of the face image in a case is shown.

  FIGS. 6A and 7A show examples of face images when a face image is taken with the left illumination unit 201a turned on. FIGS. 6B and 7B show examples of face images. b) shows an example of a face image when a face image is taken with the right illumination unit 201b turned on.

  That is, when the face of the person to be collated is in a normal state, that is, when it is not “spoofing” using a photograph or the like, the face image taken with the left illumination unit 201a turned on is shown in FIG. As shown in the figure, a shadow appears on the left side of the face, and a shadow appears on the right side of the face as shown in FIG. 6B in the face image taken with the right illumination unit 201b turned on.

  On the other hand, when the face of the person to be collated is not in a normal state, that is, in the case of “spoofing” using a photograph or the like, the right illumination unit 201a is also applied to a face image taken with the left illumination unit 201a turned on. A shadow in the same state also appears on the face image taken with the lit. In the example shown in FIGS. 7A and 7B, a shadow appears on the left side of the face as shown in FIG. 7A in the face image taken with the left illumination unit 201a turned on. As shown in FIG. 7B, a shadow appears on the left side of the face in the face image taken with the right illumination unit 201a turned on, as shown in FIG. 7B.

  This is because, when a planar image such as a photograph is presented in front of the camera unit 101, no matter which direction the light is irradiated from, the shadow on the face image taken by the camera unit 101 This is because it always depends on a planar image such as a photograph. Therefore, it is possible to determine whether or not it is “spoofing” using a planar image such as a photograph by determining the shadow state in a plurality of face images taken with light irradiated from a plurality of directions. It becomes.

  In the processes in steps S31 to S33, it is determined whether or not the shadows in the two face images photographed by sequentially turning on the two lights are normal. As an actual determination process, the shadows in the face images can be created. It is only necessary to determine whether the direction is normal. That is, in the second embodiment, it is possible to easily realize analysis processing (processing for determining whether or not a shadow is normal) for a face image captured by sequentially turning on a plurality of lights.

  As described above, according to the second embodiment, whether or not impersonation is based on whether or not the size of a face in a face image photographed by a camera is appropriate as in the first embodiment. Furthermore, whether or not the image is impersonated depending on whether or not the shadow of the face in the face image photographed by alternately lighting two illumination units arranged on the left and right of the camera is normal. Is determined.

  This makes it possible to discriminate spoofing using a flat image such as a photograph by a plurality of methods, and to prevent impersonation using a flat image such as a photograph with high accuracy and a traffic matching device. A control device can be provided.

The front view which shows the external appearance and structural example of the face collation apparatus which concerns on 1st Embodiment of this invention. It is a flowchart for demonstrating operation | movement of the face collation apparatus which concerns on 1st Embodiment. The figure which shows the positional relationship of a camera and the face of a collation target person image | photographed with the camera. The front view which shows the external appearance and structural example of the face collation apparatus which concerns on 2nd Embodiment of this invention. It is a flowchart for demonstrating operation | movement of the face collation apparatus which concerns on 2nd Embodiment. The figure which shows the example in case the face image of the collation subject illuminated by the 2nd illumination part is normal. The figure which shows an example in case the face image of the collation subject illuminated by the 2nd illumination part has the possibility of impersonation.

Explanation of symbols

  X ... distance, l ... distance, θ ... elevation angle, 1, 2 ... face matching device (traffic control device), 101 ... camera unit (imaging means), 102 ... control unit, 102a ... storage device (storage means), 103 ... Display unit 104 ... Operation unit 105 ... Distance measuring sensor unit (detection means) 107 ... Illumination unit (first illumination unit) 108 ... Signal control unit 109 ... Door 109a ... Door opening / closing mechanism unit (passage) (Permission means, passage prevention means), 201 ... second illumination unit (illumination unit), 201a ... left illumination unit, 201b ... right illumination unit

Claims (6)

In a face matching device that performs a matching process between a storage unit in which a registrant's face image is registered in advance and a face image registered in the storage unit,
Photographing means for photographing the face image of the person to be verified;
Detecting means for detecting the distance from the photographing means to the person to be collated who photographed the face image by the photographing means;
The size of the face photographed by the photographing means based on the distance from the photographing means detected by the detecting means to the person to be collated and the distance between feature points of a plurality of faces in the face image photographed by the photographing means. A judging means for judging
Determining means for determining whether or not the size of the face in the face image photographed by the photographing means determined by the determining means is appropriate;
When it is determined by this determination means that the face size in the face image photographed by the photographing means is appropriate, the face image photographed by the photographing means and the face image of the registrant registered in the storage means Execution means for executing the matching process;
When it is determined by the determining means that the face size in the face image photographed by the photographing means is not appropriate, the face image photographed by the photographing means and the registrant face image registered in the storage means Processing means for disabling collation processing;
A face collating apparatus comprising: Further, a calculating means for calculating a distance from the photographing means detected by the detecting means to a facial feature point in the face image photographed by the photographing means;
The determination means is based on the distance from the photographing means to the facial feature point calculated by the calculating means and the distance between the plurality of facial feature points in the face image photographed by the photographing means. The face collation apparatus according to claim 1, wherein the size of the photographed face is determined. The calculation means is configured to detect a face image photographed by the photographing means from the photographing means based on a distance detected by the detecting means and an angle formed by an optical axis of the photographing means with respect to a direction in which the distance is detected by the detecting means. The face collation apparatus according to claim 2, wherein a distance to a feature point of the face is calculated. In a face matching device that performs a matching process between a storage unit in which a registrant's face image is registered in advance and a face image registered in the storage unit,
Photographing means for photographing the face image of the person to be verified;
Detecting means for detecting the distance from the photographing means to the person to be collated who photographed the face image by the photographing means;
The size of the face photographed by the photographing means based on the distance from the photographing means detected by the detecting means to the person to be collated and the distance between feature points of a plurality of faces in the face image photographed by the photographing means. A judging means for judging
First determination means for determining whether or not the size of the face in the face image photographed by the photographing means judged by the judgment means is appropriate;
When it is determined by the first determination means that the face size in the face image photographed by the photographing means is appropriate, the collation target person is sequentially irradiated with light from the left side and the right side of the photographing means. Lighting means;
A shadow that appears on the face of the person to be collated when light is emitted from the left side of the photographing means by the illumination means and a shadow that appears on the face of the person to be collated when light is irradiated from the right side of the photographing means by the illumination means. Second determination means for determining whether or not is normal;
When it is determined by the second determination means that the shadow appearing on the face of the person to be collated is normal, the face image photographed by the photographing means and the face image of the registrant registered in the storage means are collated. Execution means for executing processing;
When it is determined by the first determination means that the face size in the face image photographed by the photographing means is not appropriate, or the shadow appearing on the face of the person to be collated is not normal by the second determination means If it is determined, processing means for disabling collation processing between the face image photographed by the photographing means and the registrant face image registered in the storage means;
A face collating apparatus comprising: Control of passage for passers-by based on the result of comparison between the storage means in which face images of persons permitted to pass in advance are registered and the face images of passers-by and face images registered in the storage means In the traffic control device,
Photographing means for photographing a passer's face image;
Detecting means for detecting a distance from the photographing means to a passerby who photographed a face image by the photographing means;
In the face image photographed by the photographing means based on the distance from the photographing means to the passerby detected by the detecting means and the distance between the feature points of a plurality of faces in the face image photographed by the photographing means. A judging means for judging the size of the face;
Determining means for determining whether or not the size of the face in the face image photographed by the photographing means determined by the determining means is appropriate;
When it is determined by this determination means that the face size in the face image photographed by the photographing means is appropriate, the face image photographed by the photographing means and the passage registered in the storage means are permitted. Executing means for executing a collation process with a face image of a person,
A passage permission unit that permits passage of the passer if the matching between the face image captured by the photographing unit executed by the execution unit and the face image of the person registered in the storage unit is successful;
When the collation between the face image photographed by the photographing means executed by the execution means and the face image of the person registered in the storage means fails, or the face photographed by the photographing means by the determination means If it is determined that the size of the face in the image is not appropriate, a passage blocking means for blocking the passage of the passer-by,
A traffic control device comprising: Control of passage for passers-by based on the result of comparison between the storage means in which face images of persons permitted to pass in advance are registered and the face images of passers-by and face images registered in the storage means In the traffic control device,
Photographing means for photographing a passer's face image;
Detecting means for detecting a distance from the photographing means to a passerby who photographed a face image by the photographing means;
The face in the face image photographed by the photographing means based on the distance from the photographing means to the passerby detected by the detecting means and the distance between the feature points of a plurality of faces in the face image photographed by the photographing means. A judging means for judging the size of
First determination means for determining whether or not the size of the face in the face image photographed by the photographing means judged by the judgment means is appropriate;
When it is determined by the first determination means that the face size in the face image photographed by the photographing means is appropriate, the collation target person is sequentially irradiated with light from the left side and the right side of the photographing means. Lighting means;
A shadow that appears on the face of the person to be collated when light is emitted from the left side of the photographing means by the illumination means and a shadow that appears on the face of the person to be collated when light is irradiated from the right side of the photographing means by the illumination means. Second determination means for determining whether or not is normal;
When it is determined by the second determination means that the shadow appearing on the face of the person to be collated is normal, the collation process between the face image photographed by the photographing means and the face image of the registrant registered in the storage means Execution means for executing
A passage permission unit that permits passage of the passer if the matching between the face image captured by the photographing unit executed by the execution unit and the face image of the person registered in the storage unit is successful;
When the collation between the face image photographed by the photographing unit executed by the execution unit and the face image of the person registered in the storage unit fails, the first determination unit photographs the image by the photographing unit. When it is determined that the face size in the face image is not appropriate, or when the second determination means determines that the shadow appearing on the passer's face is not normal, the traffic blocker that blocks the passer's passage Means,
A traffic control device comprising:

Images