JP2006251946A - Entry management device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an entry management device capable of permitting entry to only people with the intention of entry, without requiring operation for showing the intention of entry. <P>SOLUTION: The entry management device 50 with a tag reader 14 for reading, in a nontact manner an ID number recorded in a tag 15 is provided with a face detecting part 12 for detecting a face from images picked up by a camera 10; a facial size determining part 18 for determining the size of the face detected; a facial direction determining part 19 for determining the direction of the face detected; and an entry intention determining part 20 for determining whether there is the intention of entry according to the size and the direction of the face. When the collation of the ID number recorded in the tag 15 indicates that the ID number is normal, images of the user are picked up a plurality of times by using the camera 10. When the size of the face, for which the images are picked up meets such a requirement as getting larger, the entry intention determining part 20 determines that the user has the intention of entry. Once the user is determined to have the intention of entry, a door is opened by a door drive part 13 to permit the entry of the user. <P>COPYRIGHT: (C)2006,JPO&NCIPI

Description

  The present invention relates to an entrance management apparatus that allows only a specific person to enter a predetermined place.

  Conventionally, an entrance management device has been used to restrict a third party from accessing a room or facility that requires confidentiality. As an example, admission that allows a user to carry a tag that records an ID number for personal identification, reads the tag with a tag reader provided near the entrance, and permits or prohibits admission based on the collation result of the read ID number. Management devices are known. For example, Patent Document 1 listed below describes an admission management device that uses a non-contact tag. In this apparatus, a tag reader communicates with a non-contact tag wirelessly, and reads information such as an ID number recorded on the tag in a non-contact manner. Therefore, even if the tag is not inserted into the tag reader, the tag information is automatically read and admission permission / prohibition processing is performed only by the tag holder approaching the tag reader.

  However, in the non-contact type entrance management device as described above, since the communication distance between the tag and the tag reader is usually secured about several meters, the owner of the tag who does not intend to enter happens to enter and exit. Even when the vehicle has just passed, the tag information is read by the tag reader, the admission process is performed, and the door may be opened. This means that the door opens when it is not necessary and allows the entry of a third party, which reduces security.

  On the other hand, Patent Document 2 described later describes an entrance management apparatus provided with intention display means for displaying an intention to enter. The intention display means includes, for example, a push button switch, a touch sensor, or a voice sensor, and is provided in a reading device, a passage, or the like. In the device of this document, a person who wants to enter a specific place presses a pushbutton switch, touches a touch sensor, or inputs a voice to the voice sensor, etc. Do. On the reading device side, the door opening control is performed on the condition that the collation result of the recorded information read from the tag is correct and the intention display means outputs an intention display. According to this, when a tag holder who does not intend to enter accidentally passes the vicinity of the entrance / exit, the intention display means cannot output the intention display even if the collation result of the recorded information read from the tag is correct. If the door does not open, it is possible to avoid the problem that the door opens except when necessary.

JP 2003-6578 A JP-A-8-254050

  However, in the admission management apparatus provided with the intention display means as in Patent Document 2, some input operation for displaying the intention of admission is always required for the switch, the sensor, and the like. The convenience is not fully utilized. In addition, when the switch is accidentally pushed or the sensor is touched even though there is no intention to enter, the door opens as in the conventional case, and the security is lowered. On the other hand, if you accidentally forget to operate the intention display even though you are willing to enter, you will not be able to enter without opening the door.

  Accordingly, an object of the present invention is to provide an admission management device that can permit admission only to a person who intends to enter without performing an operation for displaying an intention to enter.

  The present invention relates to a wireless communication with a recording medium in an admission management device that collates information recorded on a recording medium possessed by a user and permits or prohibits the user from entering a predetermined place based on the collation result. Reading means for reading the information recorded on the recording medium in a non-contact manner, first determination means for judging whether the information read by the reading means is normal, and imaging for imaging the user Face detection means for detecting a face from an image acquired by the means, and second determination means for determining whether or not a face state obtained by the face detection means satisfies a predetermined condition indicating a user's intention to enter And the first determination means determines that the information read by the reading means is normal, and the second determination means determines that the face state satisfies a predetermined condition indicating the user's intention to enter In place if It is obtained by a control means for controlling to allow entrance.

  In the present invention, attention is paid to the face state obtained from the captured image of the user, and it is determined whether or not the face state satisfies a predetermined condition representing the user's intention to enter. Even if the information read from the recording medium is normal, admission is not permitted immediately, and admission is permitted when it is determined that the face condition satisfies a predetermined condition. For this reason, since the face state does not satisfy the predetermined condition for those who do not intend to enter, the entrance is not permitted and the door does not open unnecessarily. In addition, since the presence / absence of entry intention is automatically determined based on the image captured by the imaging means, the user does not need to perform any operation for displaying the intention to enter, and the convenience of the non-contact recording medium Will not be damaged. In addition, since it is not necessary to provide the intention display means, it is possible to avoid a situation where the door is opened or not opened due to an erroneous operation of a switch or the like.

  In the present invention, several methods are conceivable when the second determination means determines whether the face state satisfies a predetermined condition indicating the user's intention to enter. For example, a face size determination unit for determining a face size for each time is provided based on a plurality of face images obtained from an image captured a plurality of times by the imaging unit, and the face size for each time determined by the face size determination unit When is gradually increased, a method of determining that the predetermined condition is satisfied can be employed. Users who are willing to enter usually come straight toward the entrance and exit, and in this case, the face image captured from the front by the imaging means gradually increases, so this will help the user to enter Can be confirmed.

  Further, as another method, when the face size for each time determined by the face size determination means is compared with a predetermined reference size, and there are a predetermined number or more of face images larger than the reference size, the predetermined condition A method for determining that the condition is satisfied may be adopted. For example, even if the user retreats once before reaching the entrance / exit, if the user finally comes to the entrance / exit, the user is likely to have an intention to enter. Since the number of face images described above is equal to or greater than the predetermined number, the user's intention to enter can be confirmed.

  As another method, a face direction determination unit that determines a face direction for each time is provided based on a plurality of face images obtained from an image captured a plurality of times by the imaging unit, and the face direction determination unit determines the face direction. The face direction for each time is examined, and when there are a predetermined number or more of face images in the front direction, it may be determined that the predetermined condition is satisfied. Users who are willing to enter are usually facing the front and approaching the entrance. In this case, many face images taken from the front by the imaging means are facing the front. Can be confirmed.

  In addition, each of the above three determination methods may be adopted alone, or two or more may be combined to determine that there is an intention to enter when any one of the conditions is satisfied. .

  According to the present invention, the presence / absence of the intention to enter is determined based on the face image. Therefore, even if there is no operation for displaying the intention to enter, the person having the intention to enter is automatically identified and the entry is permitted. be able to.

  FIG. 1 is a diagram showing a schematic configuration in the vicinity of an entrance / exit gate. G is, for example, a door provided at the entrance of the facility, 10 is a surveillance camera disposed above the door G, 14 is a tag reader disposed on the side of the door G, and 15 is a non-contact possessed by the user A It is a tag (hereinafter simply referred to as “tag”). The camera 10 is composed of a visible light camera, for example, and the mounting angle is adjusted in advance so as to photograph the user A in a predetermined area from the front. The tag reader 14 includes an antenna, a transmission / reception circuit, and the like. The tag reader 14 communicates with the tag 15 wirelessly and reads information recorded on the tag 15. The tag 15 is composed of a transmission / reception circuit, a memory, and the like, and information recorded in the memory includes an ID number for identifying the user A. This ID number is assigned in advance for each user permitted to enter the facility.

  FIG. 2 is a block diagram showing an embodiment of the present invention, and a portion surrounded by a broken line shows the entrance management device 50. Reference numeral 10 denotes the above-described camera, 11 denotes an image acquisition unit that acquires an image captured by the camera 10, and 12 denotes a face detection unit that detects a face from the image acquired by the image acquisition unit 11. 13 is a door driving unit for opening and closing the door G, 14 is the above-described tag reader, 15 is the above-described tag, 16 is a display unit including a liquid crystal display, and 17 is a storage unit including a volatile memory and a non-volatile memory. . 18 is a face size determination unit that determines the size of the face detected by the face detection unit 12, 19 is a face direction determination unit that determines the direction of the face detected by the face detection unit 12, and 20 is the face size determination unit 18 and Based on the determination result of the face direction determination unit 19, an entrance intention determination unit that determines whether or not the user has an intention to enter, 21 is an ID number included in the information of the tag 15 read by the tag reader 14. An ID determination unit 22 that determines whether the admission is correct or not is a control unit 22 that controls the overall operation of the admission management device 50. In addition, each block of 12, 18-22 is actually implement | achieved as a function of a microcomputer.

  In the above configuration, the camera 10 constitutes an embodiment of the imaging means in the present invention, the tag reader 14 constitutes an embodiment of the reading means in the present invention, and the tag 15 constitutes an embodiment of the recording medium in the present invention. The face detection unit 12 constitutes an embodiment of the face detection means in the present invention, the face size determination unit 18 constitutes an embodiment of the face size determination means in the present invention, and the face direction determination unit 19 The face direction determination means in the present invention constitutes one embodiment, the ID determination section 21 constitutes one embodiment of the first determination means in the present invention, and the entrance intention determination section 20 constitutes the second determination means in the present invention. One embodiment is constituted and control part 22 and door drive part 13 constitute one embodiment of the control means in the present invention.

  FIG. 3 is a diagram showing a part of the recording area in the storage unit 17 of FIG. The storage unit 17 includes a captured image recording area 31 where a captured image captured by the camera 10 is recorded, a face size recording area 32 where a face size determined by the face size determination unit 18 is recorded, and a face direction determination. A face direction recording area 33 is provided in which the face direction determined by the unit 19 is recorded. These areas 31 to 33 are secured in a predetermined area of the volatile memory of the storage unit 17. In addition, the storage unit 17 includes a face image template recording area 34 in which several face image patterns are recorded in advance as templates, and a registered ID recording area 35 in which ID numbers of users permitted to enter are registered. Is provided. These areas 34 and 35 are secured in a predetermined area of the nonvolatile memory of the storage unit 17.

  In the face image template recording area 34, five types of face image templates of right 45 ° face image, right 20 ° face image, front face image, left 20 ° face image, and left 45 ° face image are recorded. Examples of these face image templates are shown in FIG. 4A is a face image template facing the direction of 45 ° to the right (the direction seen from the subject, the same applies hereinafter), FIG. 4B is a template of the face image facing the direction of 20 ° to the right, (c ) Is a face image template facing front, (d) is a face image template facing 20 ° to the left, and (e) is a face image template facing 45 ° to the left.

  Next, the operation of the entrance management apparatus 50 having the above configuration will be described. FIG. 5 is a flowchart showing the operation of the admission management device 50. In step S11, it is constantly monitored whether or not the record information of the tag 15 has been read by the tag reader 14. As shown in FIG. 1, when the user A approaches the tag reader 14 and the record information of the tag 15 is read by the tag reader 14 (step S11: YES), the ID determination unit 21 is included in the read record information. Whether or not the ID number has been registered is determined with reference to the registered ID recording area 35 (FIG. 3) of the storage unit 17 (step S12). As a result of the determination, if the ID number is not registered (step S12: NO), the process returns to step S11 without performing the processing after step S13, and if the ID number is registered (step S12: YES), step S13. Proceed to

  In step S <b> 13, the user A is imaged by the camera 10. The captured image data acquired by the image acquisition unit 11 is stored in the captured image recording area 31 (FIG. 3) of the storage unit 17. Next, the face detection unit 12 detects a face from the captured image (step S14). As a method for detecting a face, known methods such as a method for detecting skin color, a method for extracting a difference from a background image, and a method for extracting the likelihood of a face from pattern matching can be used.

  Subsequently, the face size is determined for the detected face (step S15), and the face direction is determined (step S16). The face size is obtained as a vertical x horizontal dimension of the detected face image. Further, in determining the face direction, the detected face image is compared with five types of face images (FIG. 4) recorded in advance in the face image template recording area 34 of the storage unit 17, and the face having the highest degree of coincidence. Adopt the face direction of the image. The face size and face direction data thus determined are stored in the face size recording area 32 and the face direction recording area 33 of the storage unit 17, respectively.

  Next, it is determined whether or not face detection by the face detection unit 12 has been performed a predetermined number of times (N times) (step S17). If the number of face detections has not reached the predetermined number (step S17: NO), the process returns to step S13, and imaging with the camera 10 is performed again. Determine the direction. These face size and face direction data are recorded in the face size recording area 32 and the face direction recording area 33 of the storage unit 17 as described above. That is, in the areas 32 and 33, the face size and the face direction are sequentially recorded for each imaging. And if imaging is performed in multiple times and the frequency | count of face detection reaches predetermined number (step S17: YES), it will move to step S18 and the presence or absence of a user's willingness to enter will be determined.

  FIG. 6 is a flowchart showing details of the determination process in step S18. First, N face size data recorded in the face size recording area 32 of the storage unit 17 are read (step S31). Then, based on the read data, it is determined whether or not the first condition “face size is gradually increasing” is satisfied (step S32). In FIG. 1, if the user A intends to enter, the user A usually approaches straight toward the door G of the entrance, so the face size of the user A captured by the camera 10 from the front is As shown in FIG. 7, it gradually increases as the number of times of imaging is increased. Therefore, in this case, the first condition is satisfied (step S32: YES), and it is determined that the user A has an intention to enter (step S36).

  On the other hand, when the first condition is not satisfied (eg, the face size fluctuates irregularly (step S32: NO)), the process proceeds to step S33 and the second condition “a face image larger than the reference size is predetermined. It is determined whether or not “there is more than the number (M)”. In FIG. 1, when the user A once retreats for some reason, the first condition is not satisfied. However, when the user A finally arrives in front of the door G, the user A Is likely to have an intention to enter. In this case, as shown in FIG. 8, it is determined that the number of face images larger than the reference size is equal to or larger than the predetermined number, the second condition is satisfied (step S33: YES), and user A has an intention to enter. (Step S36).

  If the number of face images larger than the reference size is less than the predetermined number and does not satisfy the second condition (step S33: NO), the process proceeds to step S34 and is recorded in the face direction recording area 33 of the storage unit 17. N face direction data are read out. Based on these data, it is determined whether or not the third condition “there are a predetermined number (M) or more of face images facing the front” is satisfied (step S35). In FIG. 1, if the user A intends to enter, the user A usually faces the front and approaches the door G. Therefore, the face image of the user A captured by the camera 10 from the front is as follows: As shown in FIG. 9, the front-facing ones increase. Therefore, in this case, it is determined that the number of front-facing face images is equal to or greater than the predetermined number, the third condition is satisfied (step S35: YES), and user A has an intention to enter (step S36).

  On the other hand, if the number of face images facing the front is less than the predetermined number and the third condition is not satisfied (step S35: NO), since none of the three conditions is satisfied, the user A is It is determined that there is no intention to enter (step S37). In FIG. 6, the predetermined number of values in step S33 and step S35 is the same M, but the predetermined number of values in both steps may be different.

  Returning to FIG. 5, when the entrance intention is determined as described above in step S18, a process for permitting or prohibiting entrance is performed based on the determination result. That is, when it is determined that there is an intention to enter (step S19: YES), the entrance guidance is displayed on the display unit 16 (step S20), and subsequently, the door G is opened by the door drive unit 13 (step S21). ), Allow user A to enter. FIG. 10 shows an example of a guidance message displayed on the screen 16a of the display unit 16 in step S20. The display unit 16 is provided at an appropriate location near the door G. Further, voice guidance may be performed instead of or along with display guidance. On the other hand, when it is determined that there is no intention to enter (step S19: NO), the process returns to the first step S11 without executing steps S20 and S21. In this case, the door G is not opened, and entry of the user A is prohibited.

  In this way, in this embodiment, the recorded information (ID number) of the tag 15 is collated in the first stage. If the collation is normal, the camera 10 captures the image of the user A in the second stage. To obtain a face image. Then, based on the face size and the face direction obtained from the face image of the user A, it is determined whether the face state satisfies the first to third conditions, and any one of the three conditions is satisfied. In the case, it is determined that user A has an intention to enter, and if none of the three conditions is satisfied, it is determined that user A does not have an intention to enter, and depending on these determination results, admission permission or We are banning admission. For this reason, even if the recorded information collation result of the tag 15 is normal, a user who does not intend to enter does not satisfy the above three conditions. G does not open and security is not compromised. Further, since the presence / absence of the intention to enter is automatically determined based on the image captured by the camera 10, no operation for displaying the intention to enter is required at all, and the user is freed from the troublesome operations. The In addition, since there is no need to provide a switch to display the intention to enter, the switch is accidentally pushed without the intention to enter, and the door opens or forgets to press the switch when there is an intention to enter. The situation where the door does not open can be avoided in advance.

  In the embodiment described above, three conditions of the first condition to the third condition are prepared, and it is determined that there is an intention to enter when any one is satisfied. However, when two or more of the three conditions are satisfied Alternatively, it may be determined that there is an intention to enter. Moreover, you may make it determine the presence or absence of entrance intention, using any one of 1st conditions-3rd conditions independently.

  Moreover, in the said embodiment, although the tag 15 was mentioned as an example as a recording medium, the recording medium of this invention is not restricted to this, If the recorded information can be read non-contactingly, the The form does not matter. Accordingly, not only a general contactless card but also a wristwatch type recording medium to be worn on the body or a communication device such as a mobile phone may be used.

It is the figure which showed schematic structure of the entrance / exit gate vicinity. It is a block diagram showing an embodiment of the present invention. It is the figure which showed a part of recording area in a memory | storage part. It is an example of a face image template. It is a flowchart showing operation | movement of an entrance management apparatus. It is the flowchart which showed the detail of the determination process. It is a figure which shows the face image when satisfy | filling 1st conditions. It is a figure which shows the face image when satisfy | filling 2nd conditions. It is a figure which shows the face image when satisfy | filling 3rd conditions. It is an example of the guidance message displayed on a display part.

Explanation of symbols

DESCRIPTION OF SYMBOLS 10 Camera 12 Face detection part 13 Door drive part 14 Tag reader 15 Non-contact tag 18 Face size determination part 19 Face direction determination part 20 Entrance intention determination part 21 ID determination part 22 Control part 50 Entrance management apparatus A User G Door

Claims (4)

In an admission management device that collates information recorded on a recording medium possessed by a user and permits or prohibits entry of the user to a predetermined place based on the collation result.
Reading means for performing wireless communication with the recording medium and reading information recorded on the recording medium in a non-contact manner;
First determination means for determining whether the information read by the reading means is normal;
Face detection means for detecting a face from an image acquired by an imaging means for imaging the user;
Second determination means for determining whether or not a face state obtained by the face detection means satisfies a predetermined condition representing a user's intention to enter;
The first determination means determines that the information read by the reading means is normal, and the second determination means satisfies the predetermined condition representing the user's intention to enter the face. Control means for performing control for permitting admission to the predetermined place,
An admission management device comprising: In the admission management device according to claim 1,
A face size determining means for determining a face size for each time based on a plurality of face images obtained from an image captured a plurality of times by the imaging means;
The admission management device, wherein the second determination means determines that the predetermined condition is satisfied when the face size determined by the face size determination means is gradually increased every time. . In the admission management device according to claim 1,
A face size determining means for determining a face size for each time based on a plurality of face images obtained from an image captured a plurality of times by the imaging means;
The second determination means compares the face size determined every time by the face size determination means with a predetermined reference size, and when there are a predetermined number or more of face images larger than the reference size, An admission management device that determines that a predetermined condition is satisfied. In the admission management device according to claim 1,
Based on a plurality of face images obtained from images captured a plurality of times by the imaging means, further comprising a face direction determination means for determining the face direction every time;
The second determination means examines the face direction for each time determined by the face direction determination means, and determines that the predetermined condition is satisfied when there are a predetermined number or more of face images in the front direction. A special admission management device.

Images