JP5876253B2 - Suspicious object monitoring system - Google Patents

Description

  The present invention relates to a system for determining a suspicious object such as a suspicious person in a monitoring area.

  There has been proposed a system in which a monitoring area is set in a building or its surrounding area, and a legitimate user in the area, such as a resident or an employee, is accompanied by a portable device such as an IC tag. For example, when a moving object is detected in the monitoring area, it is determined based on a radio signal emitted by the portable device whether the moving object is associated with the portable device, and whether the moving object is a legitimate user or not Can be identified.

  In the security system shown in the following Patent Document 1, when a request is received, a portable card that wirelessly transmits an identification number is used, and an employee possesses the card in the security area. When the heat ray sensor detects a moving object in the security area, the security device performs the reading operation of the identification number of the mobile card. If the registered identification number cannot be obtained, the security device determines that the moving object is an intruder. Abnormal alarm.

Japanese Patent Laid-Open No. 11-144160

  When a moving object sensor such as a heat ray sensor detects a person in the monitoring area, simply detecting the presence of a portable device such as an IC tag in the monitoring area, the detected person can wear the portable device. It cannot be judged until it is attached. In other words, the monitoring area is usually larger than the size of one person, and there may be a mobile device separately from the detected person. In this case, the person is erroneously recognized as a regular user and the suspicious person is detected. Leakage can occur. Specifically, there are cases where the mobile device exists as a lost item in the monitoring area, or there are regular users who have the mobile device in addition to persons who do not have the mobile device. Note that the background of the problem is that it is not easy to grasp the position of the IC tag with high accuracy from a remote tag reader.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a suspicious object monitoring system in which a suspicious object can be more accurately determined based on not having an IC tag or the like. .

  A suspicious object monitoring system according to the present invention is a system for monitoring the behavior of a suspicious object not accompanied by an identification wireless device, the object detecting means for detecting position information of a moving object in a monitoring area, and the identification wireless device A wireless signal detecting means for detecting the position information of the identification wireless device, and comparing the position information of the moving object and the position information of the identification wireless device to correspond to the moving object Correspondence determination means for obtaining the identification wireless device, accuracy determination means for determining the accuracy with which the moving object is associated with the identification wireless device, and the position or movement of the moving object and the accuracy with respect to the moving object. And an abnormality determining means for determining whether or not the moving object is the suspicious object. For example, the abnormality determination unit has a reference accuracy determined according to the position or movement of the moving object, and determines the moving object as the suspicious object when the accuracy related to the moving object is less than the reference accuracy. .

  In another suspicious object monitoring system according to the present invention, a history of position information is recorded for each moving object, tracking processing means for tracking the moving object, and a predetermined specific action related to the moving object are recorded. Action detecting means for detecting based on a history of the position information, and the abnormality determining means, when the specific action by the moving object is detected, the accuracy with respect to the moving object is the specific action If it is less than the reference accuracy set for, the moving object is determined as the suspicious object.

  In the suspicious object monitoring system according to still another aspect of the present invention, the tracking processing unit further records a history of the accuracy for each moving object to be tracked, and the abnormality determining unit includes the moving object based on the moving object. When a specific action is detected, the suspicious object is determined by comparing the accuracy history of the moving object with the reference accuracy.

  In the suspicious object monitoring system according to another aspect of the present invention, the abnormality determination unit may be configured such that the accuracy is less than the reference accuracy even when the accuracy when the specific action by the moving object is detected is less than the reference accuracy. The specific action is permitted if the accuracy is equal to or higher than the set relaxation standard accuracy and there is a time in the accuracy history that is equal to or higher than the reference accuracy.

  According to the present invention, the accuracy of suspicious object determination is improved by determining the suspicious object in consideration of the probability (accuracy) that the moving object is carrying the identification wireless device. In addition, by changing the reference accuracy serving as a threshold for the determination according to the position and movement of the moving object to be determined as a suspicious object, an accurate suspicious object determination in consideration of the risk for each action is realized.

It is a block diagram which shows the structure of the outline of the security system comprised including the suspicious person monitoring system which is embodiment of this invention. It is a typical top view of the monitoring area where a suspicious person monitoring system is installed. It is a block diagram which shows the schematic structure of the detection apparatus which comprises a suspicious person monitoring system. It is a general | schematic flowchart explaining the process in a link process part. It is a general | schematic flowchart of an example of the link maintenance determination process in a link process part. It is a general | schematic flowchart which shows an example of the determination process of a link level. It is a general | schematic flowchart which shows the other example of the determination process of a link level. It is a block diagram which shows the schematic structure of the monitoring apparatus which comprises a suspicious person monitoring system. It is a general | schematic flowchart of the determination process at the time of the detection of an entrance event. It is a general | schematic flowchart of the determination process at the time of the detection of a stay event and an outer periphery movement event. It is a general | schematic flowchart of the determination process at the time of detection of an entrance event.

  Hereinafter, a suspicious person monitoring system 2 according to an embodiment of the present invention (hereinafter referred to as an embodiment) will be described with reference to the drawings. FIG. 1 is a block diagram illustrating a schematic configuration of a security system including a suspicious person monitoring system 2. The security system includes a suspicious person monitoring system 2, a communication network 4, and a security center 6.

  The suspicious person monitoring system 2 is installed in the property to be monitored. For example, the suspicious person monitoring system 2 monitors the site on the outer periphery of the monitored building. More specifically, for example, a detached house is the monitoring target, and the garden / approach of the house is the monitoring area. The suspicious person monitoring system 2 can also be used for indoor monitoring. For example, an entrance hall or a lobby of a condominium, office building, hotel or the like can be used as a monitoring area.

  The suspicious person monitoring system 2 monitors a moving object (person) in the monitoring area in a state where a regular user (hereinafter referred to as a user) such as a resident or employee of the monitored object may exist in the monitoring area. That is, unlike the security system that sets the security start when the user is not present in the monitoring area such as at night, or performs the security release operation when the user can exist, the suspicious person monitoring system 2 is guarded. Basically, an operation such as release is not required, and monitoring can be performed even in a state where the user can exist such as during the day. The suspicious person monitoring system 2 monitors a person including a user in the monitoring area. When a suspicious person who has entered the premises or a suspicious person who has entered the building is detected, an image of the monitoring area together with an abnormality notification signal is sent to the security center 6. send. Upon receiving the abnormality report, the security center 6 can take appropriate measures such as monitoring the situation based on the sent image and dispatching the security guard.

  The suspicious person monitoring system 2 includes a monitoring device 10, a detection device 12, and an operation indicator 14. The monitoring device 10 is connected to the detection device 12 and the operation indicator 14 so that they can communicate with each other via wireless / wired. In addition, it is connected to the security center 6 via the communication network 4.

  One or a plurality of detection devices 12 are installed in the property to be monitored. The monitoring device 10 makes an abnormality determination by integrating the detection results from the detection device 12, and performs processing such as reporting to the security center 6 based on the result. Although details will be described later, for example, a wireless tag 16 is provided in advance as an identification wireless device that can identify whether the user is a user, for example, a regular user (hereinafter referred to as a user) such as a resident or an employee of the monitored property. Carry this in the surveillance area. The suspicious person monitoring system 2 detects the position of the moving object in the monitoring area and receives the wireless signal emitted from the wireless tag to acquire the identification information and position.

  FIG. 2 is a schematic plan view of a monitoring area where the suspicious person monitoring system 2 is installed. The site on the outer periphery of the building 300 is set as a monitoring area. A wall 302 such as a hedge or a fence is provided at the boundary between the site and the site, and a site entrance / exit 304 is provided in a part thereof. The detection device 12 is installed on, for example, an outer wall of a building 300, and a laser pulse 306 is irradiated from a sensor unit 20 (described later) to a monitoring area to detect a person 308 that is a moving object. Further, a question signal is issued from the reader unit 22 to the wireless tag 16, and the reader unit 22 detects the wireless tag 16 based on a response signal from the wireless tag 16 and acquires its position.

  The suspicious person monitoring system 2 collates the position of the moving object and the position of the wireless tag to perform an association process (link process) for obtaining a wireless tag corresponding to the moving object, and also performs accuracy (link level) according to the collation result. ) The link level is an index indicating the probability that the moving object is carrying the wireless tag. That is, if the link level is high, it indicates that the person associated with the wireless tag has a high probability of carrying (possessing) the wireless tag. (It cannot be said that the probability is high.) A moving object that cannot be associated with a wireless tag is given the lowest link level indicating that no link exists.

  The suspicious person monitoring system 2 uses this link level to authenticate whether or not each moving object is a user carrying a wireless tag, and for example, determines a moving object whose link level is lower than a reference level as a suspicious object. I do. In this determination, different reference levels can be set according to the position or movement of the moving object. For example, if the garden is an area where customers and delivery personnel cannot pass, the garden can be set to a higher reference level than the approach that they are supposed to use. Further, the suspicious person monitoring system 2 is configured to record the position history of each moving object, track the moving object, and monitor the occurrence of various determination target events taking movement into consideration. When the determination target event set based on the position and movement is detected, the link level of the moving object that is the subject of the event is compared with the reference level set in accordance with the content of the event, and as described above An intrusion abnormality is determined.

  Next, the configuration of the detection device 12 will be described with reference to FIG. The detection device 12 includes a sensor unit 20, a reader unit 22, a camera 24, a detection processing unit 26, a storage unit 28, a notification unit 30, and a LAN communication unit 32.

  The sensor unit 20 has a function of detecting position information of a person in the monitoring area. In this embodiment, the sensor unit 20 is a laser distance sensor, and scans a monitoring area with a laser beam to detect the position of an object that reflects the laser beam. For example, the sensor unit 20 includes a near-infrared laser light source having a wavelength of about 890 nm, reflects a laser pulse from the light source with a scanning mirror that is driven to rotate at a constant speed, changes its emission direction, and covers the entire monitoring area. Irradiate sequentially. Further, the sensor unit 20 receives the laser pulse reflected by the object and generates distance measurement data up to the object. The scanning of the laser beam is repeated at a predetermined cycle (for example, 200 milliseconds) with respect to an angle range (for example, 180 ° in the horizontal direction) facing the monitoring area with the sensor unit 20 as the center. Ranging is performed using a time-of-flight method (TOF method: Time of Flight) at predetermined angle steps (for example, 0.25 °) within the scanning angle range, specifically from emission of laser pulses to reception of light. Is measured, and the distance to the object reflecting the laser is calculated from the time and the speed of light. The distance measurement data is represented by a scanning angle (azimuth) and a distance, and the position of the object viewed from the detection device 12 is given by the distance measurement data.

  The sensor unit 20 is not limited to the laser distance sensor described above, and can be similarly configured with, for example, a radar sensor that uses radio waves such as microwaves and millimeter waves, and an ultrasonic sensor (sonar sensor) that transmits and receives ultrasonic waves. It is. The image sensor is used to extract a person based on the background difference and inter-frame difference, determine the direction of the person from the horizontal coordinate on the image, and reverse perspective from the vertical coordinate in the image obtained by photographing the surveillance area obliquely from above. Ranging data can also be generated by estimating the distance to a person by conversion.

  The reader unit 22 receives a wireless signal emitted from the wireless tag 16 given to the user, and detects position information of the wireless tag 16. The wireless tag 16 stores unique identification information (tag ID), and when a question signal from the reader unit 22 is received, a response signal including the tag ID is transmitted as a wireless signal. In this embodiment, the wireless tag 16 is configured by using an active RFID (Radio Frequency IDentification) tag provided with a battery, and communicates with the reader unit 22 in a relatively wide monitoring area of the suspicious person monitoring system 2. Make it possible. As the wireless tag 16, for example, a UHF band such as 900 MHz band or a microwave band frequency such as 2.45 GHz band can be used. Note that some passive RFID tags that rectify radio waves from the reader and use as a power source can communicate for several meters. If the monitoring area is relatively small, passive tags can be used. is there.

  The reader unit 22 sends a question signal to the monitoring area. Then, the reader unit 22 receives the response signal emitted from the wireless tag 16 in response to this, and extracts the tag ID included in the response signal.

  The reader unit 22 has a configuration for detecting position information of the wireless tag 16 based on the received response signal. For example, the detection device 12 detects, as the position information of the wireless tag 16, the azimuth in which the wireless tag 16 is seen from the reader unit 22. As an example of a configuration for that purpose, the reader unit 22 includes a plurality of antennas each having a directivity with a relatively narrow beam width (for example, 10 °) and arranged with the beam directions shifted from each other. Sends out a question signal and receives a response signal. The reader unit 22 switches the directivity so that the transmission / reception is sequentially performed in all directions (360 °), and the reading of the wireless tag 16 in all directions is completed every predetermined cycle (for example, 200 milliseconds). The reader unit 22 acquires the reception intensity of the response signal of the wireless tag 16 for each directivity angle sequentially set in all directions, summarizes the reception intensity in all directions for each tag ID, and performs detection processing as reception data for one cycle. To the unit 26.

  In the above example, the beam direction is changed in steps corresponding to the beam width. That is, if the beam width is about 10 °, the directivity angle is set in steps of about 10 ° while providing some overlap between the beams. However, the step of changing the beam direction can be set smaller than the beam width, whereby the direction in which the wireless tag 16 exists can be obtained at an angle smaller than the beam width.

  For link processing described later, it is necessary to define in advance the relationship between the position based on the distance measurement data and the position (orientation) based on the tag data. Preferably, the sensor unit 20 and the reader unit 22 are arranged close to each other in the detection device 12 so that the visual axes of the sensor unit 20 and the reader unit 22 substantially coincide with each other. Thereby, the scanning angle of the sensor unit 20 and the directivity angle of the reader unit 22, that is, the angle (azimuth) in the distance measurement data and the angle (azimuth) of the reception data can be expressed in a common coordinate system.

  The antenna provided in the reader unit 22 can have another configuration in which the direction of directivity can be changed. For example, a configuration in which the directivity angle is changed by rotating a single antenna having strong directivity in a certain direction, or a configuration in which the directivity angle is changed by electronic control, such as a phased array antenna, is possible. Further, a transmitting antenna and a receiving antenna may be provided, and only one of them may have a sharp directivity, and the other may have a wide directivity such as omnidirectionality.

  The reader unit 22 obtains the azimuth as the position information of the wireless tag 16 in the above example, but may be configured to obtain the position in the monitoring area. For example, the reader unit 22 can obtain the distance to the wireless tag 16 based on the time difference from the transmission of the inquiry signal to the reception of the response signal, and can determine the position on the two-dimensional coordinate by the azimuth and the distance. In addition, the structure which determines the position on a two-dimensional coordinate by the principle of triangulation using the reception data obtained by the reader part 22 of each detection apparatus 12 arrange | positioned in multiple places is also possible. In this case, it is preferable to design the processing sharing between the detection device 12 and the monitoring device 10 so that the processing for detecting the position of the wireless tag 16 is performed by the monitoring device 10.

  The camera 24 captures the monitoring area and acquires an image. For example, an image signal of the monitoring area is generated using a CCD image sensor, a CMOS image sensor, or the like. An image photographed by the camera 24 is sent from the suspicious person monitoring system 2 to the security center 6 together with an abnormality notification signal, and a monitor of the security center 6 uses it for grasping the situation. In addition, the user can confirm the monitoring area from the image on the operation display 14.

  The storage unit 28 stores various programs executed by the detection processing unit 26 and data necessary for the programs. In particular, the storage unit 28 stores the tag ID information (registered tag data) of the wireless tag 16 given to the user and the distance measurement data (reference distance measurement data) acquired by the sensor unit 20 in the absence of a person. Remember.

  The detection processing unit 26 is configured using a microprocessor or the like, and according to a program to be executed, the distance measurement data acquired by the sensor unit 20, the tag data detected by the reader unit 22, and the position of the wireless tag 16. Process. Specifically, the detection processing unit 26 functions as a moving object detection unit 40, a tag detection unit 42, a link processing unit 44, and an accuracy determination unit 46.

  The moving object detection means 40 and the sensor unit 20 constitute object detection means for detecting the position of the moving object in the monitoring area. The moving object detection means 40 extracts a group of scanning angles (distance change group) in which distance changes of a predetermined distance or more have occurred in the distance measurement data input from the sensor unit 20 as corresponding to the moving object. Then, when the size of the moving object grasped from the distribution of the distance and the scanning angle in the distance change group is about a person, the moving object is determined to be a person, and the direction and distance at which the person is detected are determined as the person. The position of The position (azimuth, distance) of the person is, for example, the azimuth and distance of the center of the distance change group.

  When a plurality of distance change groups are extracted, person determination and position detection are performed for each group. A person ID is assigned to each detected person. The moving object detection means 40 determines whether the persons are the same, for example, by determining the size similarity and positional relationship between the person detected in the previous scan and the person detected in the current scan. For example, regarding the positional relationship, the validity of the moving speed derived from the moving distance and elapsed time of the person can be used as a material for determining whether or not they are the same person.

  It should be noted that the azimuth / distance of the person may be expressed by other representative values such as a scanning angle / distance distribution range of the distance change group and an average value.

  The tag detection means 42 together with the reader unit 22 constitutes a wireless device detection means for detecting the presence and position of the wireless tag 16 in the monitoring area. The tag detection means 42 receives received data that summarizes the azimuth and reception intensity at which the tag ID is detected for each tag ID from the reader unit 22. The tag detection unit 42 refers to the registered tag data in the storage unit 28, and when the detected tag ID is registered, the detected direction of the wireless tag 16 corresponding to the direction having the highest reception intensity for each tag ID. And

  Note that the orientation of the wireless tag 16 may be expressed by other representative values such as a distribution range of directivity angles, an average value, and a median value.

  The link processing means 44 obtains the wireless tag 16 corresponding to the moving object by comparing the position information of the person detected by the moving object detection means 40 with the position information of the wireless tag 16 detected by the tag detection means 42. Functions as correspondence determination means. In the present embodiment, when the wireless tag 16 is detected, the link processing unit 44 compares the orientation (person's orientation) included in the information for specifying the position of the person with the orientation of the wireless tag 16, and the coincidence thereof. Is recognized, the person is associated with the wireless tag 16 (link establishment). On the other hand, when no match is recognized or when the wireless tag 16 is not detected, no association is made (link failure).

  FIG. 4 is a schematic flowchart for explaining processing in the link processing means 44. The moving object detection means 40 identifies the position of the person (S50), and if the person is not a person being tracked, that is, a person newly detected by the current scanning of the sensor unit 20 (in S52, “ In the case of “No”, a person ID is assigned to the person (S54), and link setting is attempted (S58 to S68). On the other hand, if the person being tracked, that is, a person who has already been assigned a person ID (“Yes” in S52), the link processing unit 44 determines whether or not a link exists for the person ID. That is, it is determined whether it is associated with the wireless tag 16 (S56). If tracking is in progress but no link exists (“No” in S56), link setting is attempted (S58 to S68).

  If the link does not exist (“No” in S52 and S56), the link processing means 44 has detected the wireless tag 16 by the tag detection means 42 (in the case of “Yes” in S58), and When the link is not set for the tag ID of the wireless tag 16 (in the case of “Yes” in S60), the difference amount (azimuth deviation amount) between the orientation of the person and the orientation of the wireless tag 16 is calculated (S62). ). If the amount of heading deviation is within the predetermined threshold Th1 (in the case of “Yes” in S64), the tag ID of the wireless tag 16 is associated with the person ID assigned to the person who is the target of the link process. Thereby, a new link is set (S66).

  On the other hand, when the wireless tag 16 is not detected in the monitoring area (in the case of “No” in S58), or when the wireless tag 16 is detected but there is no unlinked one (in the case of “No” in S60). In addition, when the azimuth deviation amount exceeds the threshold Th1 (in the case of “No” in S64), it is a case where the link cannot be set or it is inappropriate to set. Therefore, in these cases, the link processing unit 44 does not set a link for the person to be processed as a link failure (S68).

  The link processing unit 44 performs a link maintenance determination process S70 when a link is set for the person to be processed in the previous process (in the case of “Yes” in S56). The link maintenance determination process S70 is a process that takes into account that the detection error of the wireless tag 16 is relatively large. For the person who has once set a link and the wireless tag 16, the criteria for determining the coincidence are relaxed in the subsequent link processing. This avoids the frequent occurrence of broken links.

  FIG. 5 is a schematic flowchart of an example of the link maintenance determination process S70. When the link exists, the link processing unit 44 replaces the processes S58 to S68 when the link does not exist, instead of the link maintenance determination process S70. Execute. Specifically, in the link processing unit 44, the tag ID of the wireless tag 16 to be determined for link maintenance has been detected by the tag detection unit 42 (in the case of "Yes" in S80 and S82), and the link maintenance determination target Azimuth | direction misalignment amount about a person and the wireless tag 16 is calculated (S84). If the azimuth deviation amount is within the threshold Th2 set to Th2> Th1 (in the case of “Yes” in S86), it is determined that a link is established between the person to be determined for link maintenance and the wireless tag 16. Then, the association between the two is maintained (S88). On the other hand, when the wireless tag 16 is not detected in the monitoring area (in the case of “No” in S80), or when the wireless tag 16 is detected but the tag ID that is the target of the link maintenance determination is not detected (S82). In the case of “No” at the time), and when the amount of heading deviation exceeds the threshold Th2 (in the case of “No” in S86), it is a state that should be a link failure, and is a determination target to be maintained. The association is released (S90). For example, the threshold Th1 can be set to 5 ° and Th2 can be set to 8 °.

  In addition, when the detection direction of the person or the wireless tag 16 is expressed as a range as described above, it depends on whether or not the angle ranges overlap each other, or if they are not overlapped, how far the nearest ends are separated from each other. Determination processing S64 and S86 for the azimuth deviation amount is performed.

  In the configuration for obtaining the two-dimensional coordinates of the wireless tag 16 described above, the link processing is performed based on the coincidence of the two-dimensional coordinates instead of the coincidence of the directions. That is, the determination processes S64 and S86 are performed with the distance between the position of the person given by the azimuth and distance and the position represented by the two-dimensional coordinates of the wireless tag 16 as the amount of deviation.

  The link processing means 44 performs the above-described link processing for each person detected in the monitoring area.

  The accuracy determination unit 46 functions as a accuracy determination unit that determines a link level, which is a probability that a person is carrying the wireless tag 16. The accuracy determination unit 46 determines the link level based on the result of association between the person and the wireless tag 16 in the link processing unit 44. The link level is set in multiple stages.

  FIG. 6 is a schematic flowchart showing an example of a link level determination process. In this example, the accuracy determination means 46 determines the link level based on the correspondence between the person and the wireless tag 16, that is, the link establishment persistence. It can be said that the sustainability of link establishment is higher, for example, as the duration is longer. In addition, the stability of the link establishment state is one manifestation of the sustainability. For example, the followability in which the link establishment state is maintained following the movement of the person can be used as an indicator of stability. The greater the movement of the person, the more stable and high the link with the wireless tag 16 that follows it. The link level can be set.

  In the example of FIG. 6, the link level is defined in three stages of “2”, “1”, and “0” in descending order, and the level “0” means that no link exists. Specifically, the accuracy determination means 46 sets the link level to “0” when the link is not established (“No” in S100) (S102). The link is established (in the case of “Yes” in S100), and the state continues for a predetermined time or more (in the case of “Yes” in S104), and the followability is high. When the amount of change in the azimuth of the person since the establishment is equal to or greater than a predetermined value (“Yes” in S106), the highest link level “2” is set (S108). On the other hand, if either the condition that the duration is equal to or greater than the threshold value or the condition that the followability is high is not satisfied (in the case of “No” in S104 or S106), an intermediate link level “1” is set ( S110). Note that the duration can be measured in an arbitrary unit, for example, the number of times of link processing.

  Here, as an example in which the link level is set higher as the link establishment lasts, a simple example in which the link level has three stages has been shown. However, for example, the duration determination time is set to 2 or more and the duration is longer. A higher link level can be set, and similarly, the higher the azimuth followability, the higher the link level can be set. In the example of FIG. 6, two conditions of the link level “2” in the link establishment state (in the case of “Yes” in S100) are satisfied: the duration condition (S104) and the followability condition (S106). However, it may be determined whether the level is set to “2” or “1” only in any one of these two conditions. Moreover, it is good also as an intermediate | middle level of the level when not satisfy | filling both the level when satisfy | filling both the link levels when only one of conditions S104 and S106 is satisfy | filled. Further, in the configuration in which the two-dimensional coordinates of the person are detected, the followability can be determined using the position change amount (movement distance) instead of the azimuth change amount.

  FIG. 7 is a schematic flowchart showing another example of the link level determination process. In this example, the accuracy determination unit 46 sets different link levels according to the detection accuracy of the wireless tag 16. Specifically, when the detection accuracy is high, a higher link level can be set than when the detection accuracy is low. The detection accuracy of the wireless tag 16 can be determined based on the magnitude of the peak value of the received intensity in the received data, the azimuth distribution of the received intensity, and the like. Conceptually, if the peak position and shape are clear, the detection accuracy is estimated to be high. For example, the lower the peak, that is, the weaker the response signal from the wireless tag 16, the more difficult it is to distinguish between the peak and noise, and the peak position is more susceptible to noise. The detection accuracy is estimated to be low. In addition, a plurality of peaks may appear in the received data corresponding to a certain wireless tag 16. For example, when the response signal transmitted from the wireless tag 16 is reflected by another object and reaches the reader unit 22, or when the side lobe is present in addition to the main lobe in the antenna directivity of the reader unit 22. . For example, when the difference between the maximum peak value and the second peak value in a plurality of peaks is small, the probability that the maximum peak is a true peak is low, and the detection accuracy is estimated to be low.

  The detection accuracy of the wireless tag 16 can also be determined based on consistency in a temporal change in detection direction. Conceptually, it is estimated that the detection accuracy is low when the time change of the detection direction deviates greatly from the general way of human movement.

  In the example of FIG. 7, the link level is defined in four stages of “3”, “2”, “1”, and “0” in descending order. When the link is not established (“No” in S120), the accuracy determination unit 46 sets the link level to “0” (S122). The link is established (in the case of “Yes” in S120), and the state continues for a predetermined time or more (in the case of “Yes” in S124), and the direction of the person from the time when the link is established for the first time. If the amount of change is equal to or greater than a predetermined value (“Yes” in S126), the highest link level “3” is set (S128). On the other hand, when either the condition that the duration is equal to or greater than the threshold value or the condition that the followability is high is not satisfied (in the case of “No” in S124 or S126), the accuracy determination means 46 has a predetermined tag detection accuracy. If the threshold value is higher than the threshold value (S130), if it is equal to or higher than the threshold value (if “Yes” in S130), the medium link level “2” is set (S132). (In the case of “No” in S130) A low link level “1” is set (S134).

  Note that the determination result that the sustainability of link establishment is high is based on results, and in this case, even if the detection accuracy is low, it can be said that the detection direction of the wireless tag 16 is reliable. Therefore, in the example shown in FIG. 7, the link level is determined by giving priority to the sustainability of the link establishment over the tag detection accuracy. That is, the presence / absence of persistence is determined (S124, S126). If the persistence exists, a high level “3” is set as the link level at the time of persistence (S128). On the other hand, if the persistence does not exist, the detection accuracy is determined (S130), and based on the determination result, the medium level “2” that complements the link level sequence in the range below the link level “3” at the time of the sustain. And the low level “1” is set (S132, S134).

  Also in the example of FIG. 7, the same modification as described in the example of FIG. 6 can be performed.

  Further, in the process of FIG. 7, one of the determination of the duration (S124) and the determination of the followability (S126) may be omitted. Further, both of the determinations (S124, S126) may be omitted, and a plurality of levels in the link establishment state may be set only by the tag detection accuracy. On the other hand, when it is determined that persistence exists in the determination of the duration (S124) and the determination of followability (S126), the detection accuracy of the tag is further determined, and the detection accuracy is less than the threshold value. Level “3” may be set, and level “4” may be set when the level is equal to or greater than the threshold.

  For example, when the reader unit 22 is configured to detect a tag with high accuracy based on the transmission / reception time difference described above, the determination threshold value of the tag detection accuracy in the accuracy determination unit 46 can be increased. In this case, the determination that the tag detection accuracy is high can be more important than the determination that the persistence exists, and the link level that prioritizes the tag detection accuracy over the persistence, contrary to the example of FIG. Can be set. For example, in a configuration in which the link level is set in five stages of 0 to 4, when the tag detection accuracy is high, level “4” is obtained if there is further sustainability of the link, and level “3” if there is no sustainability. When the tag detection accuracy is low, the level is “2” if the link establishment persistence exists, the level “1” if the persistence does not exist, and the link failure is set to the level “0”.

  Further, in addition to or instead of determination of sustainability and tag detection accuracy, it is possible to determine the amount of azimuth deviation and set the link level. In this configuration, the smaller the azimuth misalignment is, the higher the link level is set.

  In this embodiment, the detection device 12 is configured to be connected to the monitoring device 10 via a LAN (Local Area Network), and the LAN communication unit 32 provided in the detection device 12 is a communication interface with the monitoring device 10.

  In addition, the notification unit 30 is a speaker or the like that outputs a warning at the time of abnormality detection described later. The detection processing unit 26 receives an instruction from the security center 6 or the monitoring device 10 via the LAN communication unit 32 and operates the notification unit 30.

  The operation indicator 14 includes a monitor that displays a monitoring image taken by the camera 24 of the detection device 12. For example, the user of the suspicious person monitoring system 2 can check the monitoring area with an image at any time on the operation display 14. Further, the operation indicator 14 is provided with input means such as buttons, switches, and numeric keys for performing various operations on the monitoring device 10 and the detection device 12, and a speaker for outputting a voice guide or an alarm when an abnormality is detected. it can.

  Next, the configuration of the monitoring device 10 will be described with reference to FIG. The monitoring device 10 includes a monitoring processing unit 150, a storage unit 152, a LAN communication unit 154, and a reporting unit 156.

  The LAN communication unit 154 is a communication interface with the detection device 12. On the other hand, the reporting unit 156 is a communication interface with the security center 6 via the communication network 4.

  The storage unit 152 stores various programs executed by the monitoring apparatus 10 and data necessary for the programs. In particular, the storage unit 152 stores map information 160 relating to the monitoring area and tracking information 162 that manages the actions (movement trajectory and the like) for each person detected by the detection device 12. Further, the position represented by the polar coordinate system (azimuth and distance) centered on the detection device 12 in the distance measurement data by each detection device 12 is, for example, on a map represented by a two-dimensional coordinate system common to the monitoring area. Information for converting to coordinates is stored.

  The map information 160 is information that represents the entire area of the monitoring area, information that represents the arrangement of each partial area that is set in the monitoring area according to the monitoring attribute, and touches the building of the monitored property among the boundaries of the monitoring area Information indicating the position of the part and the building entrance (hereinafter referred to as the building entrance), and the part of the boundary of the monitoring area that touches the outside of the site and the position of the entrance from the outside of the site to the monitoring area (hereinafter referred to as the site entrance) Contains information to indicate. As a partial area set as a monitoring area, for example, an entrance area where a person other than a user such as a guest or a delivery person is expected to exist, and a normal monitoring area where a person other than the user is basically not expected to exist Can be set. Specifically, the entrance area includes a route (approach) from a site entrance to a building entrance, and may further include a neighboring area.

  The monitoring processing unit 150 is configured using a microprocessor or the like, and functions as a tracking processing unit 164, a behavior detection unit 166, and a determination processing unit 168 according to a program to be executed.

  The tracking processing means 164 receives human detection data generated by the moving object detection means 40 sequentially from the detection device 12. If the tracking information 162 is not recorded in the storage unit 152 for the person ID existing in the person detection data, the tracking information 162 for the person ID is newly created. On the other hand, when the tracking information 162 is already recorded for the person ID existing in the human detection data, the tracking information 162 is updated.

  The tracking processing means 164 converts the position (azimuth and distance) of the person in the human detection data into the coordinates on the map. The coordinates are recorded as tracking information 162 together with other information such as a person ID, detection time, link information (link level and tag ID).

  The behavior detection unit 166 detects various determination target behaviors (events) based on the tracking information 162. In the present embodiment, an entrance event, an exit event, a stay event, an outer peripheral movement event, an entrance event, and an exit event are defined as determination target actions. Each of these events will be described later.

  When the determination processing unit 168 detects the determination target action, the determination processing unit 168 executes a determination process according to the type of the action. In particular, when a specific action that requires an abnormality determination as to whether or not a person is a suspicious person is detected, whether or not the user is a user is identified with reference to the link level of the person, and an intrusion abnormality is determined. Since the degree of risk when a person is a suspicious person varies depending on the type of specific action, the required security level is also different. Therefore, the link level required for each determination is set as the reference level (reference accuracy).

  The specific behavior includes an entrance act for detecting an abnormal entry into the building, an entrance act for detecting a suspicious act within the premises (monitoring area), a staying act, and an outer periphery moving act. Admissions have a higher security level than other specific actions. Here, Lvh> Lvm> Lvl, where the reference link level of the admission act is Lvh, the reference link level of other specific actions is Lvm, and the lower limit of the link level set in the link establishment state is Lvl. . For example, in the example of FIG. 7, Lvh = 3, Lvm = 2, and Lvl = 1 are set.

  In addition, the determination processing unit 168 does not make the abnormality uniformly if the link level at the time of detection of the specific action does not satisfy the standard, but is sufficiently large during the previous tracking even if the standard does not satisfy the standard slightly. If the link level is obtained, the specific action is allowed. Thereby, a stable determination can be realized by absorbing detection errors of a person or the wireless tag 16.

  Hereinafter, the detection condition of each event and the content of the determination process at the time of detection will be described.

[Admission event]
For a newly detected person (new person ID), a case where the appearance position is a boundary with the outside of the site is detected as the event.

  FIG. 9 is a schematic flowchart of the determination process when an entry event is detected. If the link level set for the detected person is equal to or higher than the reference link level Lvm for the specific action other than the above-described admission action (“Yes” in S200), the person's entry action is permitted. (S202).

  When the link level of the person is less than Lvm, it is determined whether or not the entrance is from the site entrance, that is, whether the appearance position corresponds to the site entrance location (S204). When entering from the entrance of the site (“Yes” in S204), there is a possibility of being a visitor such as a customer or a delivery person. An entry into the entrance area is permitted (S206). In this case, the determination processing unit 168 activates the stay timer for the person and starts measuring the certain time. On the other hand, if the entrance is from other than the entrance of the site (“No” in S204), it is determined that the suspicious person has entered, and the outer periphery intrusion abnormality is determined (S208). The tracking processing unit 164 continues the tracking process even after the determination by the determination processing unit 168.

  Incidentally, since the tracking information 162 is newly created when an entrance event is detected, the above-described determination processing in consideration of the link level during tracking is not performed.

[Exit event]
A case where the person being tracked disappears from the monitoring area and the position immediately before it is a boundary with the outside of the site is detected as the event.

  When the event occurs, the tracking processing unit 164 ends the tracking. Although the 162 regarding the person ID can be deleted, in this embodiment, the tracking information 162 is held at least until the abnormality is recovered for the person determined to be abnormal during tracking.

[Still event]
A case where the person being tracked with temporary permission set exists in the entrance area for a certain time is detected as the event.

  FIG. 10 is a schematic flowchart of determination processing at the time of detecting a stay event and at the time of detecting an outer peripheral movement event described later. If the link level of the person at the time of the event detection is equal to or higher than Lvl (in the case of “Yes” in S220) and the link level is higher than Lvm during tracking (S222) At "Yes"), admission permission is set (S224). On the other hand, when the link level is less than Lvl when the event is detected (in the case of “No” in S220), or when the link level has never satisfied the reference level Lvm in both the current time and the past (in S222) In the case of “No”), the outer periphery intrusion abnormality is determined (S226).

  In other words, if a person who has been temporarily permitted has stayed for more than a certain period of time, it may be a person who has an unauthorized purpose such as picking or hiding, so it is basically recognized as a suspicious person and has a perimeter intrusion abnormality. (S226). On the other hand, if the reference level Lvm is not satisfied at the time of event detection but has been satisfied in the past, the link level may be less than Lvm due to a detection error of the detection device 12. Therefore, in order to prevent misreporting due to detection errors of the detection device 12, the reference level for the link level required at the present time is relaxed to Lvl, the relaxation criterion level Lvl is satisfied at the present time, and the original standard for staying at the present time or in the past If the level Lvm is satisfied, the admission permission setting is made and staying is permitted (S224). However, even if there is a timing equal to or higher than Lvm in the past, it is not permitted until the current link level is less than the relaxation reference level Lvl (in the case of “No” in S220).

  Although the link level history is referred to in the above determination, the determination may be made based on whether or not the reference level Lvm is satisfied at the time of detection of the stay event without referring to the history.

[Outside movement event]
When the person being tracked moves from the entrance area to the regular monitoring area, the event is detected.

  Since the person who has been temporarily permitted may include persons with an unfair purpose such as vacant nests and latencies, the movement to the regular monitoring area is basically recognized as a suspicious person and is regarded as an abnormal intrusion into the outer periphery (S226). On the other hand, when the event level is the reference level Lvm at the time of event detection or in the past, in order to prevent false alarms due to detection errors of the detection device 12, the entrance permission setting is set to allow movement to the normal monitoring area (S224). However, even if there is a timing equal to or higher than Lvm in the past, it is not permitted until the current link level is lower than the reference level Lvl (in the case of “No” in S220).

  Note that the outer peripheral movement event may also be determined based on whether or not the reference level Lvm is satisfied at the time of detection of the event without referring to the link level history.

[Entrance event]
A case where the person being tracked disappears from the monitoring area and the position immediately before it is a boundary with the building area is detected as the event.

  FIG. 11 is a schematic flowchart of the determination process when an entrance event is detected. In the determination processing means 168, the link level of the person at the time of the event detection is equal to or higher than the reference level Lvm of the admission permission setting (in the case of “Yes” in S240), and the link level is the reference level Lvh of the admission action during tracking. When the above state exists (in the case of “Yes” in S242), the entrance permission setting is performed (S244). On the other hand, when the link level is lower than the relaxation reference level Lvm at the time of the event detection (in the case of “No” in S240), or when the link level has never satisfied the reference level Lvh at both the present time and the past (In the case of “No” in S242), the building intrusion abnormality is determined (S246).

  As a general rule, admissions with a high security level should not be permitted except by persons who can be surely possessed of the wireless tag 16, and as described above, a higher standard level Lvh than other specific acts is set. Yes. However, in order to prevent false alarms due to detection errors of the detection device 12, the reference level for the link level required at the present time is relaxed to the reference level Lvm for the admission permission setting, and the present relaxation level Lvm is satisfied at the present time, and the current or past If the original standard level Lvh for admission is satisfied, admission is permitted (S244). However, even if there is a timing equal to or higher than Lvh in the past, no entry is permitted until a person whose link level is currently less than the relaxation standard level Lvm, that is, a level where admission is not permitted (in the case of “No” in S240).

  Although the link level history is referred to in the above determination, the determination may be made based on whether or not the reference level Lvh is satisfied at the time of detection of the entrance event without referring to the history.

  In the above-described embodiment, the wireless tag 16 receives a question signal from the reader unit 22 and issues a response signal. However, the wireless tag 16 may spontaneously transmit at a certain period or when vibration is detected.

  In the above embodiment, the detection device 12 is responsible for person detection, tag detection, link processing, and link level determination, and the monitoring device 10 is responsible for person tracking, event detection, and event determination. It is not limited to what is shown in the form.

  As described above, in the present invention described in the embodiment, the positions of the person detected in the monitoring area and the wireless tag 16 are collated to associate the person with the wireless tag 16. Thereby, it is possible to accurately grasp which person in the monitoring area has which wireless tag 16. Further, in the present invention, the suspicious person determination is not performed based on the presence / absence of the association between the person and the wireless tag 16 but is performed in consideration of the probability that the person has the wireless tag 16. In general, the position detection of the RFID tag is affected by the influence of the surrounding environment such as the presence or absence of an object that reflects radio waves such as walls and planting, but the variation increases, but in the present invention, by considering the link level representing the accuracy, A more suitable determination is possible.

  Further, when a person takes a specific action that should be determined as abnormal, it is determined whether or not the person is a suspicious person using the link level. At this time, by setting the link level of the determination standard according to the type of the specific action, it becomes possible to determine the suspicious person according to the security level required for each specific action.

  2 Suspicious person monitoring system, 4 communication network, 6 security center, 10 monitoring device, 12 detection device, 14 operation indicator, 16 wireless tag, 20 sensor unit, 22 reader unit, 24 camera, 26 detection processing unit, 28, 152 Storage unit, 30 notification unit, 32, 154 LAN communication unit, 40 moving object detection unit, 42 tag detection unit, 44 link processing unit, 46 accuracy determination unit, 150 monitoring processing unit, 156 reporting unit, 160 map information, 162 tracking Information, 164 tracking processing means, 166 action detecting means, 168 determination processing means.

Claims (3)

A suspicious object monitoring system for monitoring the behavior of a suspicious object not accompanied by an identification wireless device,
Object detection means for detecting position information of the moving object in the monitoring area;
A radio detection means for receiving a radio signal emitted by the identification radio and detecting position information of the identification radio;
Correspondence determination means for collating position information of the moving object and position information of the identification wireless device to obtain the identification wireless device corresponding to the moving object;
Accuracy determining means for determining the accuracy with which the moving object is associated with the identification wireless device;
A tracking processing means for recording the position information history for each moving object and tracking the moving object;
Action detecting means for detecting a predetermined specific action related to the moving object based on a history of the position information;
Abnormality determining means for determining whether or not the moving object is the suspicious object based on the position or movement of the moving object and the accuracy related to the moving object;
Have
When the specific action by the moving object is detected, the abnormality determination unit determines that the moving object is the suspicious object if the accuracy related to the moving object is less than a reference accuracy set for the specific action. Judging,
Suspicious object monitoring system. The suspicious object monitoring system according to claim 1 ,
The tracking processing means further records the accuracy history for each moving object to be tracked,
The abnormality determining unit determines the suspicious object by comparing the accuracy history of the moving object with the reference accuracy when the specific action by the moving object is detected;
Suspicious object monitoring system. The suspicious object monitoring system according to claim 2 ,
The abnormality determination means is equal to or more than a relaxation reference accuracy set lower than the reference accuracy, even if the accuracy when the specific action by the moving object is detected is less than the reference accuracy, and Allowing the specific action if there is a time in the history of the accuracy equal to or higher than the reference accuracy,
Suspicious object monitoring system.

Images