JP2013030079A - Mobile monitoring system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To surely detect a suspicious person who does not have a portable unit by making a person within a monitor area corresponding to a portable unit.SOLUTION: A mobile monitoring system which identifies whether a mobile is accompanied with a portable unit comprises object detection means which detects a location of the mobile in the monitor area, receiving means which receives a radio signal transmitted by the portable unit, azimuth detection means which detects a location or an azimuth of the portable unit from the received radio signal, corresponding processing means which compares the location or azimuth of the mobile with that of the portable unit and, in the case where a degree of matching of the location or azimuth meets a first criterion, makes the mobile corresponding to the portable unit, and chasing means which records the detected location of the mobile together with a result of corresponding and chases the mobile. Regarding the mobile made corresponding to the portable unit, the corresponding processing means maintains the relevant corresponding in the case where a degree of matching of the location or azimuth with the portable unit meets a second criterion looser than the first criterion.

Description

  The present invention relates to a mobile body monitoring system that identifies whether or not a mobile body is attached to a mobile device in a monitoring area and performs abnormality determination according to whether the mobile device is attached.

  Conventionally, a person who has entered a monitoring area is detected by a security sensor, and a portable device such as an IC tag possessed by a legitimate user is read by a reader, and the detected person is a legitimate user or a suspicious person. A system to identify this is proposed.

  For example, in Patent Literature 1, when a heat ray sensor detects a moving object in a warning area, an operation for reading an identification number of a mobile card possessed by an employee is performed, and this is performed when a registered identification number cannot be read. A security system that judges a moving object as an intruder and gives an alarm is disclosed.

Japanese Patent Laid-Open No. 11-144160

  The security system described in Patent Document 1 can recognize whether or not a portable device exists within the detection range when a certain heat ray sensor detects a person, but does the detected person truly possess a portable device? It is not possible to determine whether or not. If the detection range of the person and the portable device is narrow enough for one person to enter, it can be said that the detected person possesses the portable device, but if the detection range is wide, the portable device is located at a position away from the person. Even if exists, the person is mistakenly recognized as a regular user. In addition, even when a person who does not have a portable device is mixed among a plurality of persons, a situation that cannot be extracted as a suspicious person may occur.

  The present invention has been made to solve the above problems, and an object of the present invention is to provide a moving body monitoring system capable of accurately determining whether or not the detected moving body is accompanied by a portable device.

In order to achieve the above object, a moving body monitoring system according to the present invention is a moving body monitoring system for identifying whether a moving body is attached to a portable device, and for detecting an object position in a monitoring area. Means, receiving means for receiving a radio signal emitted by the portable device, azimuth detecting means for detecting the position or azimuth of the portable device from the received wireless signal, the position of the mobile body and the position of the portable device Or, compare the azimuth specified by the position of the mobile body with the azimuth of the portable device, and if the degree of coincidence of the position or the azimuth satisfies the first standard, the mobile body and the portable device are handled. And a tracking unit for recording the detected position of the moving body together with the result of the association and tracking the moving body, and the correspondence processing unit is associated with the portable device. Has been For mobile maintains the association if the position or degree of matching the orientation with the portable device satisfies a loose second reference than the first reference, and wherein the.
According to such a configuration, since the mobile object is associated with the mobile device on the basis of the matching condition between the position or orientation where the mobile object is detected and the position or orientation where the mobile device is detected, the mobile object It is possible to accurately determine whether it is attached. In addition, the relationship between the person and the portable device once associated with each other and the association criterion is loosened to make it difficult to cancel the association. However, it is possible to accurately determine whether the mobile device is attached to the mobile object.

  Further, in the above configuration, when the correspondence processing means has a moving amount that is not associated with the portable device, the amount of deviation of the position or the direction from the portable device is within a first reference value. For a mobile object associated with the portable device and associated with the portable device, the portable object is provided if the deviation from the portable device is within a second reference value that is greater than the first reference value. Maintain the association with the machine.

Further, in the above configuration, the response processing unit may be configured so that the position of the portable device or the direction changes following the change in the position accompanying the movement of the moving body, and the moving body and the Associate with a portable device.
According to such a configuration, not only the position or orientation of the mobile body and the portable device coincide with each other, but also the association is performed on the condition that the mobile device is moving following the movement of the mobile body. It is possible to prevent other portable devices not attached to the mobile phone from being erroneously associated with each other, and to make an auxiliary determination with high accuracy.

Further, in the above configuration, when the specific action by the mobile body is detected according to the history of the position of the mobile body, and the specific action by the mobile body is detected, the portable device is associated with the mobile body And an abnormality determining means for determining that the object is a suspicious object.
According to such a configuration, it is possible to appropriately determine whether or not the moving body that has taken the specific action is a suspicious person.

  Further, in the above configuration, the object detection means is a distance measurement sensor, scans a monitoring area with an exploration signal, generates distance measurement data indicating a distance to an object in each direction, and generates the distance measurement data from the distance measurement data. The position of the moving body in the monitoring area is detected. Alternatively, the object detection means is an image sensor that extracts a moving body from an image obtained by photographing the monitoring area, and detects the position of the moving body in the monitoring area based on the position of the extracted moving body on the image. .

  According to the present invention, by associating the mobile body with the mobile device based on the coincidence of the position or orientation while taking the detection accuracy into consideration, it is determined whether or not the detected mobile body is attached to the mobile device. Can be accurately identified.

It is a schematic whole block diagram of the suspicious person monitoring system of this embodiment. It is a figure which shows the outline of the flow of the matching process and abnormality determination process of a monitoring process part. It is a schematic diagram for demonstrating the example of a matching process. It is a flowchart which shows a matching process. It is a flowchart which shows a matching process. It is a flowchart which shows a part of matching process in a modification.

  As an embodiment of the mobile monitoring system of the present invention, a suspicious person monitoring system that detects a suspicious person who has illegally entered a building site and issues an alarm is described below.

  The suspicious person monitoring system according to the present embodiment is applied to, for example, monitoring of a service user's house, and detects an intrusion of a suspicious person into a garden corresponding to the house's premises to give an abnormality alarm. The user is given a portable device in advance, and possesses the portable device when entering and leaving the house. The suspicious person monitoring system detects a person who has entered the garden from outside the site, and specifies its position. Further, the presence of the portable device is detected by receiving a wireless signal from the portable device, and the received wireless signal is analyzed to detect the position or orientation where the portable device is present. Then, the position where the person is detected is compared with the position or orientation where the portable device is detected. If the degree of coincidence satisfies the first standard, the person is considered to have the portable device, and the authorized user Recognize that On the other hand, if the degree of coincidence does not meet the first criterion or the portable device is not detected, the detected person is recognized as a suspicious person who does not have a portable device, and for example, staying in the garden or building An abnormal alarm is output for the act of entering the building.

  The position or azimuth of the portable device is calculated by analyzing the arrival azimuth and reception intensity of the radio signal emitted by the portable device on the reader side. However, the direction of arrival and the received intensity are not uniformly stable because they are affected by various environmental factors such as reflection of radio signals by the walls of the house. This instability is directly related to the detection accuracy of the position or orientation of the portable device, and further affects the above-described determination result of the association between the person and the portable device. Therefore, even if the person who actually owns the mobile device moves within the monitoring area, if the position or orientation detection error temporarily increases, the association between the person and the mobile device fails, and the normal use There is a risk of misidentifying a person as a suspicious person.

Therefore, in the suspicious person monitoring system according to the present embodiment, for the person once associated with the portable device, only the relationship with the same portable terminal is used to loosen the criteria for subsequent association, and the association is inadvertent. Try not to come off. In other words, if the degree of coincidence of the position or orientation with the portable device already associated with the person satisfies the second criterion that is looser than the first criterion, the association with the portable device is maintained.
Thus, in order to perform the association based on the coincidence of the position and orientation of the person and the portable device in consideration of the detection accuracy of the portable device, it is accurately determined whether or not the detected person has the portable device. And the accuracy of suspicious person determination can be improved.

FIG. 1 is a schematic overall configuration diagram of the suspicious person monitoring system of the present embodiment.
The suspicious person monitoring system mainly includes a wireless tag 10, a reader 20, a distance measuring sensor 30, and a monitoring device 40, and at least a part of the site around the building is a monitoring area. When a suspicious person who has entered the building site or the building is detected, the remote security center 60 is notified of the occurrence of an abnormality via a communication line 50 such as the Internet network. The security center 60 that has received the abnormality report implements emergency measures such as dispatching security personnel to the site. In addition, if a camera (not shown) is separately provided in the suspicious person monitoring system so that an image at the time of occurrence of an abnormality can be confirmed by the security center 60, the monitor of the security center 60 can accurately check the situation at the site. I can grasp.

  The wireless tag 10 is a portable device that is given in advance to a user who is permitted to enter or leave the building, and is carried when the building that is the subject of monitoring enters or exits. The wireless tag 10 performs wireless communication using radio waves with the reader 20, and transmits a wireless signal including unique identification information (hereinafter referred to as tag ID) for each wireless tag 10 stored in advance. For the wireless communication system, RF-ID (Radio Frequency IDentification) technology is used. The wireless tag can use various frequency bands depending on the application, and there are various driving methods such as an active type and a passive type. Although the usage frequency and the driving method are not limited, in the present embodiment, in order to be used for monitoring in a building site, the UHF band suitable for relatively long-distance communication such as several meters to several tens of meters or 2. A response signal including a tag ID is output with respect to an interrogation signal from the reader 20 by using a frequency of 4 GHz band and using its own battery as a drive source. Note that a vibration detection sensor may be provided in the wireless tag 10 so that a wireless signal is transmitted only during vibration detection, that is, during movement.

  The reader 20 detects the wireless tag 10 existing in a predetermined detection range. The detection range depends on the signal strength of the radio wave transmitted by the wireless tag 10 and the reader 20. The reader 20 receives a response signal issued by the wireless tag 10 in response to the inquiry signal transmitted to the detection range, and extracts a tag ID that identifies the wireless tag 10 included in the response signal. The detection range of the reader 20 needs to include the monitoring area so that the wireless tag 10 possessed by the person in the monitoring area can be detected. Since the reach range of the question signal changes depending on the environment such as the nature of the object existing in the space, the detection range is set in advance to be wider than the monitoring area.

The reader 20 includes an antenna having directivity of a constant angle (for example, 10 °), and transmits a question signal and receives a response signal while electronically switching the directivity angle of the antenna at fixed time intervals. The reader 20 reads the wireless tag 10 over the entire monitoring area at a predetermined cycle (for example, 200 msec), acquires the reception intensity of the response signal for each directivity angle, and outputs it to the monitoring device 40 as tag detection data. Tag detection data is created for each detected tag ID. The angle defining the directivity angle is, for example, one of the horizontal directions around the visual axis of the reader 20 as a positive angle and the other as a negative angle.
As a configuration for obtaining the orientation of the wireless tag 10, a configuration in which a directional antenna is rotated to switch a directional angle, a plurality of directional antennas are mounted, and the weight of each antenna is controlled to control the directional angle. It is possible to substitute a known method such as a configuration for switching between.

  The distance measuring sensor 30 scans the monitoring area with the laser beam and detects the position of the object reflecting the laser beam. The distance measuring sensor 30 sequentially projects, for example, near infrared rays having a wavelength of about 890 nm to the entire monitoring area while reflecting the light using a scanning mirror that is driven to rotate at a constant speed. Further, the reflected light reflected by the object is received, and distance measurement data is generated as a result of the laser light irradiation. The scanning is performed at a predetermined cycle (for example, 200 msec) with respect to the range of the monitoring area (for example, 180 °). The distance measuring sensor 30 measures the distance to the object to be reflected from the time difference required from the irradiation of the laser light to the reception of the reflected light at every predetermined scanning angle (for example, 0.25 °) (so-called Time of Flight method). The measurement distance for each scanning angle is output to the monitoring device 40 as distance measurement data for one cycle. The distance measurement data is represented by an angle and a distance, and indicates the position of the reflected object with the distance measurement sensor 30 as a viewpoint. For example, one of the horizontal directions around the visual axis of the distance measuring sensor 30 is a positive angle and the other is a negative angle.

  Here, it is necessary to define in advance the relationship between the “position” indicated by the distance measurement data and the “azimuth” indicated by the tag detection data for the association processing in the monitoring device 40 described later. In the present embodiment, the reader 20 and the distance measuring sensor 30 are installed in the vicinity so that the respective visual axes coincide, and the scanning angle of the distance measuring sensor 30 and the directivity angle of the reader 20, that is, the angle in the distance measurement data and the tag detection data. Are preliminarily adjusted so that the angles are expressed in a common coordinate system. Preferably, the reader 20 and the distance measuring sensor 30 are arranged in the same casing.

  The suspicious person monitoring system according to the present embodiment includes a pair of readers 20 and distance measuring sensors 30. However, the suspicious person monitoring system includes a plurality of readers and distance measuring sensors according to the shape of the region to be monitored. . For example, when monitoring the four surroundings of a building, at least two sets of readers and distance measuring sensors are provided on the diagonal of the site. When installed on the wall of a building, a total of four readers and ranging sensors are installed on each surface.

The monitoring device 40 is connected to the reader 20 and the distance measurement sensor 30 wirelessly or by wire, and executes the monitoring process by inputting the tag detection data transmitted from the reader 20 and the distance measurement data transmitted from the distance sensor 30. When an abnormality is detected, the remote security center 60 is notified of the abnormality.
The monitoring device 40 mainly includes a storage unit 42, a notification unit 44, a notification unit 46, and a monitoring processing unit 48.

  The storage unit 42 includes a ROM, a RAM, and the like. In addition to data and programs necessary for various types of processing by the monitoring processing unit 48, a tag ID of a wireless tag previously assigned to a user, reference distance measurement data used for person detection Remember. The reference distance measurement data is distance measurement data obtained by the distance measurement sensor 30 scanning the monitoring area in the absence of a moving body. The storage unit 42 has an area in which the tracking data regarding the detected person is stored in a rewritable manner, and the tracking data is created, updated, and deleted by the monitoring processing unit 48 as needed.

The reporting unit 44 is a communication interface for transmitting an abnormality signal to the security center 60 via the communication line 50.
The notification unit 46 includes a buzzer, a speaker, a warning lamp, and the like, and is a means for notifying the surroundings that an abnormality has occurred and threatening a suspicious person. Note that the notification unit 46 may be provided in the reader 20 and the distance measuring sensor 30.

  The monitoring processing unit 48 receives distance measurement data from the distance sensor 30 and tag detection data from the reader 20, detects a person in the monitoring area, detects a wireless tag, and associates the detected person with the wireless tag. Processing, person tracking processing, and abnormality determination processing are executed. Moreover, each part of the monitoring apparatus 40 is controlled. Further, the reader 20 and the distance measuring sensor 30 are controlled, and synchronous control is performed so that the scanning cycle of the distance measuring sensor 30 coincides with the detection cycle of the reader 20.

FIG. 2 is a diagram showing an outline of the flow of the association processing and abnormality determination processing of the monitoring processing unit.
The monitoring processing unit 48 inputs the distance measurement data obtained from the distance measurement sensor 30 (S200), and performs difference extraction by comparing with the reference distance measurement data in the storage unit 42. The distance is the same as if it shows the same body, and a group of scanning angles (distance change group) in which the distance change from the reference distance measurement data is more than a predetermined value is extracted, and a new object appears at that position. recognize. The size of the appearing object is estimated from the angle range and the distance value of the distance change group, and the presence of a person is recognized when the size of the appearing object is about a person. Subsequently, the monitoring processing unit 48 detects a position where a person exists (S210). Specifically, the human detection data is generated with the central angle of the distance range indicating the person as the azimuth (person azimuth) of the person and the average of the distances corresponding to the angle range as the distance to the person. When a plurality of distance change groups are extracted from the distance measurement data, person determination and position detection are performed for each group.

  In parallel with the processes S200 to S210, the monitoring processor 48 inputs the tag detection data obtained from the reader 20 (S220), and determines whether a pre-registered tag ID is included. When the registered wireless tag 10 is detected, the orientation distribution indicating the reception intensity for each directivity angle is analyzed for the tag detection data of each tag ID, and the wireless tag has the peak directivity angle with the largest reception intensity. It is detected as an existing orientation (tag orientation) (S230). Since the directivity angle has a certain angle range, the central angle of the peak directivity angle is used for the tag orientation.

  Next, the monitoring processing unit 48 compares the position of the person obtained from the person detection data with the direction of the wireless tag 10 obtained from the tag detection data, and the person direction included in the information indicating the position of the person and the tag of the wireless tag If the degree of coincidence with the azimuth satisfies a predetermined standard, the person is associated with a set of wireless tags (S240). Thereby, it is recognized that the detected person possesses the wireless tag. If the degree of coincidence between the person orientation and the tag orientation does not satisfy a predetermined standard, it is not recognized that the person has the wireless tag at that time. However, the predetermined reference used for the association determination is not always constant, and for the person and the wireless tag that are already associated in the previous determination, the matching criterion used for the determination of the person and the wireless tag that are not associated with each other Also, the association is determined using a loose standard.

  FIG. 3 is a schematic diagram for explaining an example of the association processing. The example of FIG. 3 represents a situation when the person P having the wireless tag T moves from point a to point b and point c. When a person P carrying the wireless tag T appears at a point a in the monitoring area, the person orientation of the person P is compared with the tag orientation of the wireless tag T from the distance measurement data and tag detection data input at this time, An angle deviation amount θa is obtained. If the deviation amount θa is within a first reference value (for example, 10 °), the person P and the wireless tag T are associated (link establishment). If the wireless tag is not detected at the point a, or if the shift amount θa between the person orientation and the tag orientation exceeds the threshold, the wireless tag is not associated with the person P.

  Next, at the point b, the detection accuracy of the direction of the RFID tag T is lowered, and the amount of deviation θb from the direction of the person P is larger than the amount of deviation θa. For example, when the deviation amount θb is 12 °, since the original reference value exceeds 10 ° which is the first reference value, the association between the person P and the wireless tag T is not established. However, since the association between the person P and the wireless tag T is already established at the point a, the person P and the wireless tag T are compared with a second reference value (for example, 15 °) lower than the first reference value. Here, since the deviation amount θb is within the second reference value, the association between the person P and the wireless tag T is established, and the association is maintained (link maintenance). Similarly, at the point c, the deviation amount θc between the person P and the wireless tag T is compared with the second reference value. On the other hand, when another person Q or another wireless tag G is detected at the same time, the association between the person Q and the wireless tag G or wireless tag T is determined by the first reference value. Further, the association between the person P and the wireless tag G is also determined by the first reference value.

  Returning to FIG. 2, the monitoring processing unit 48 assigns a person ID to a person who first appears in the monitoring area, and records and accumulates the detected position change of the person in the storage unit 42 for each person ID as tracking data ( S250). The tracking data is data in which a person ID, a person detection position (angle, distance), presence / absence of association, tag ID and detection direction (angle) of the associated wireless tag, and time stamp information (detection time) are associated with each other. It is.

  Referring back to the example of FIG. 3, first, as tracking data related to the person P appearing at the point a in the monitoring area, the assigned person ID and detection position, the tag ID and detection orientation of the associated wireless tag T, detection The time is recorded in the storage unit 42. When the person P moves to the point b, the detection position at the point b, the tag ID and detection direction of the wireless tag T, and the detection time are added to the tracking data related to the person ID of the person P. The same applies when moving to point c. Whether the detected person is the same person is determined by the similarity in size with the person detected last time and the positional relationship (for example, the validity of the moving speed derived from the moving distance and elapsed time of the person). To do.

  The monitoring processing unit 48 analyzes the distance measurement data and the tag detection data sequentially input to update the tracking data, and based on the detection position and the tracking elapsed time from the first detection in the tracking data, A predetermined event occurrence is detected (S260). Event detection conditions include, for example, the same person staying in the monitoring area (staying), moving from the monitoring area to the inside of the building (entrance), leaving the monitoring area (exit), etc. It is. The stay event is detected when the same person is detected in the monitoring area for a predetermined time or longer. The entrance event is detected when the person being tracked is not detected and the detection position immediately before the disappearance of the person is near the boundary with the building. The exit event is detected when the person being tracked is no longer detected and the detection position immediately before the disappearance of the person is near the boundary with the outside of the site.

  In the present embodiment, occurrence of a stay event and an entrance event among the above events is recognized as a specific action that requires suspicious person determination. When detecting the specific behavior of the person, the monitoring processing unit 48 determines whether the event occurrence is caused by the behavior of the suspicious person or the authorized user and makes an abnormality determination (S270). Specifically, the association state of the wireless tag 10 is confirmed by referring to the tracking data regarding the person updated at the time of occurrence of the event or immediately before. If the wireless tag 10 is associated, that is, if the tag ID is recorded with respect to the person ID, it is recognized and permitted as an action by an authorized user. If the wireless tag 10 is not associated, it is recognized as an action by a suspicious person, an abnormality is determined, and an alarm process such as an abnormality report to the security center 60 by the report unit 44, an abnormality report or threat by the report unit 46 is performed. In addition, when an entrance event or an exit event occurs, the monitoring processing unit 48 ends the tracking of the person, and erases the tracking data related to the person after a certain time (for example, 10 seconds). When the entrance event is an intrusion abnormality by a suspicious person, the tracking data is retained even after the tracking is completed until a separate abnormality reset signal is input.

Next, the operation of associating a person with a wireless tag will be described in detail.
4 and 5 are flowcharts showing the association processing in the monitoring processing unit. The monitoring processing unit 48 executes association processing every time distance measurement data is input from the distance sensor 30.

The monitoring processing unit 48 analyzes the distance measurement data, performs a person detection process, and determines whether a person has been detected (S400). When a plurality of persons are detected, the subsequent processing is performed for each person. If no person is detected, the association process is terminated. At this time, if the person being tracked is not detected for a certain period of time, the tracking of the person is terminated.
If a person is detected, the presence position is detected for each person (S402), and it is determined by referring to the tracking data whether the person is being tracked (S404). If the person to be processed is not the person being tracked, a new person ID is assigned to the person to be processed (S406), and the process proceeds to the next process S410. In the case of the person being tracked, it is determined whether or not the link has already been established with any of the wireless tags (S408). If the link has not been established, the process proceeds to S410.

  In process S410, the monitoring processor 48 determines whether or not a registered wireless tag is detected based on the input of the tag detection data. If a wireless tag has been detected, it is determined whether there is a wireless tag that has not been established with another person being tracked among the detected wireless tags (S412). If there are wireless tags that have not been established, the tag orientations for those wireless tags are detected (S414). Subsequently, the azimuth deviation amount is calculated by comparing the person azimuth of the person to be processed and the tag azimuth (S416), and the azimuth deviation amount with the wireless tag having the smallest azimuth deviation amount is within the first reference value Th1. It is determined whether it exists (S418). If the amount of heading deviation is within the reference value Th1, a link is established by associating the person ID of the person to be processed with the tag ID of the wireless tag (S420), and tracking data relating to the person ID is created and recorded. (S422). Then, it is determined whether or not there is an unprocessed person among the detected persons (S424). If there is an unprocessed person, the process returns to process S402. If all the persons have been processed, the input distance measurement data is processed. The association process ends. If there is a person being tracked that has not been detected for a certain period of time, the tracking of the person is terminated.

  On the other hand, when the wireless tag is not detected (No in S410), when all the detected wireless tags have already been linked with another person (Yes in S412), the orientation of the person to be processed and the wireless tag If the amount of deviation exceeds the reference value Th1 (No in S418), tracking data relating to the person ID is created and recorded without being associated with the wireless tag (S422).

Returning to the process S408, if the person to be processed is being tracked and a link has been established with a specific wireless tag (Yes in S408), a link maintenance determination process is performed (S426).
FIG. 5 is a flowchart showing the operation of the link maintenance determination process. The monitoring processor 48 determines whether a wireless tag has been detected based on the input of tag detection data (S500). When the wireless tag is detected, the tag ID of the link target wireless tag associated with reference to the tracking data regarding the person to be processed is extracted, and the link target wireless tag is extracted from the detected wireless tag. It is determined whether there is a tag (S502).

  If a link-target wireless tag is detected, the tag orientation related to the wireless tag is detected (S504), and the orientation deviation amount is calculated by comparing the person orientation with the tag orientation (S506). If the azimuth deviation amount is within the second reference value Th2 that is greater than the first reference value Th1, the association state is maintained (S512), the link maintenance determination process is terminated, and the process proceeds to process S422 in FIG. The tracking data regarding the person ID is updated.

  In addition, when the link target wireless tag is not detected (No in S500 or No in S502), and when the amount of heading deviation exceeds the reference value Th2 (No in S508), the link target wireless tag is associated. It is determined whether the interrupted state has continued for a certain time (for example, 1 sec) (S510). If the fixed time has not elapsed, there is a possibility that the wireless tag to be linked cannot be detected temporarily, or there is a possibility that the error in the detection direction is large, so the association state is maintained (S512). However, if it continues for a certain time, it recognizes that the person to be processed does not have a wireless tag, and cancels the association (S514). Then, the link maintenance determination process is terminated, and the process proceeds to the process S422 in FIG. 4 to update the tracking data related to the person ID.

  As described above, according to the suspicious person monitoring system of the present embodiment, the person is associated with the wireless tag based on the coincidence between the direction in which the person detected in the monitoring area exists and the direction in which the wireless tag exists. Therefore, it can be determined with high accuracy that the detected person truly possesses the wireless tag. Therefore, even when a plurality of persons and a plurality of wireless tags are detected, it is possible to accurately grasp whether each person has a wireless tag and which wireless tag each person has. A person who does not have a tag can be reliably identified. In addition, regarding the person and the wireless tag that are once associated, the association reference is relaxed in the subsequent tracking process, and the association is difficult to be released. Accordingly, even if the measurement error of the detection direction of the wireless tag temporarily increases, it is possible to prevent the person from being inadvertently released and being recognized as a suspicious person. Alternatively, the same effect can be obtained when the measurement error of the person detection position temporarily increases.

In the association processing in the above embodiment, the person and the wireless tag are associated with each other if the person orientation matches the tag orientation. In this case, the dead angle of the person X as viewed from the distance measuring sensor 30 is different. If the person Y is present, the wireless tag possessed by the person Y is erroneously associated with the person X, and there is a possibility that the suspicious person's staying abnormality cannot be detected.
A modification of the above embodiment that solves this problem will be described below.

(Modification)
Also in the suspicious person monitoring system of the modified example, the overall configuration and most of the processes are the same as those in the above embodiment, but a part of the association process is different. In the association process in the modified example, the person and the wireless tag are associated with each other on the condition that the person has moved a certain amount or more while maintaining the state where the detection orientation of the person and the wireless tag is matched. Specific processing will be described below.

The association process in the modification is obtained by replacing the process S420 with the processes S600 to 606 in FIG. 6 in the flowchart of FIG. 4 described in the above embodiment.
When the amount of azimuth deviation between the person who has not established the link and the wireless tag is within the first reference value Th1 threshold (Yes in S418 in FIG. 4), the monitoring processing unit 48 determines the person ID of the person to be processed. A temporary link state is established by temporarily associating the tag ID of the wireless tag (S600), and then it is determined whether or not the detected orientation of the person has changed by a certain angle (for example, 20 °) or more by referring to the tracking data (S20). S602). That is, it is determined whether or not the person is moving in a direction in which the azimuth changes as viewed from the distance measuring sensor 30 and the reader 20.

If the person orientation has changed by a certain angle or more, it is determined whether or not the temporary link state has continued during that time (S604). When the person orientation changes due to the movement of the person carrying the wireless tag, the wireless tag moves together with the person. At this time, the change of the tag direction of the wireless tag should follow the same locus following the change of the person direction. Therefore, in process S604, it is confirmed whether the movement of the wireless tag follows the movement accompanied by the change in the direction of the person.
If the temporary link state continues, the person is associated with the wireless tag, and the link is established (S606). On the other hand, if the person orientation has not changed by more than a certain angle yet (No in S602) or if the temporary link state has not continued (No in S604), the link is not established at this point. When the processes S600 to S606 are finished, the process proceeds to process S422 in FIG. 4 to update the tracking data related to the person ID. The provisional link state release condition is that the provisional association process in step S600 is not updated for a certain period of time.

  Thus, in the association process of the modified example, the person is moved in the direction in which the azimuth changes as viewed from the reader, and the wireless tag is moved following the movement of the person. The wireless tag is associated. Therefore, for example, even if there is a person with a wireless tag near the outside of the monitoring area or a wireless tag is left in the monitoring area, the person who does not actually own the wireless tag It is possible to prevent the wireless tag from being associated. As a result, it becomes possible to realize an advanced suspicious person monitoring system that suppresses detection omission of suspicious persons.

As mentioned above, although the mobile body monitoring system concerning this invention has been demonstrated using the above-mentioned embodiment and modification, the application range of this invention is not limited to this.
For example, in the above embodiment and the modification, the central angle of the angle range forming the distance change group is used as the person orientation, but the angle range itself forming the distance change group may be the person orientation. In this case, the association is determined based on whether the detected tag direction of the wireless tag is included in the angle range of the person or whether the amount of deviation from the angle range of the person is within the reference value.

In the above-described embodiment and modification, the center of the directivity angle of the reader when the wireless tag is detected is used as the tag orientation, but the angle range of the directivity angle itself may be used as the tag orientation. In this case, the association is determined based on whether the angle indicating the human azimuth is included in the directivity angle at which the wireless tag is detected or whether the amount of deviation from the angular range of the directivity angle is within the reference value.
Furthermore, the angle range itself that forms the distance change group may be the person orientation, and the angle range of the directivity angle itself may be the tag orientation. In this case, the association is determined based on whether the person orientation and the tag orientation overlap by a certain amount or more, or whether the shift amount is within a reference value.

  Further, in the association processing in the above embodiment and the modification, the matching of the orientation of the person and the wireless tag is determined, but the determination may be made using the matching of the position of the person and the wireless tag. In this case, since it is necessary to specify the location of the wireless tag, for example, the distance to the wireless tag is calculated using the reception strength of the response signal and the time difference from the transmission of the question signal to the reception of the response signal. The position of the wireless tag is calculated from the azimuth and distance. Alternatively, it is possible to use a plurality of readers whose detection ranges overlap each other, analyze response signals from wireless tags with the same tag ID, and specify the position of the wireless tag using triangulation technology. Even when the association processing is performed based on the position matching, it is obvious that at least the orientation matching is included in the matching determination condition.

In the above-described embodiment and modification, the distance measuring sensor using laser light is used as means for detecting the position and orientation of a person. However, for example, a sensor using ultrasonic waves, microwaves, or the like can be used. In addition, it is possible to employ a configuration in which an image sensor that detects a person by image analysis is used to convert the position of the person on the image into a position in the real space and determine the position matching. In this case, the position of the person may be detected by specifying the azimuth by the horizontal coordinate on the image, and estimating the distance by inverse perspective transformation of the vertical coordinate.
Further, instead of a configuration in which a change in the position of a person is tracked and an event for determining a suspicious person is extracted, an event may be extracted based on a change in the direction of the person and a suspicious person may be determined. In this case, the sensor for detecting the person may be configured to detect only the direction of the person.

  Moreover, in the said embodiment and modification, although it was set as the structure which performs a suspicious person determination by the presence or absence of the correlation when a specific action is detected, the level set to a correlation and the correlation previously set according to the specific action The suspicious person may be determined depending on whether the level is satisfied. The association level is set according to, for example, the continuity of the link establishment state, the magnitude of the azimuth deviation amount, the direction detection accuracy of the wireless tag, and the like. If the link establishment state is long, it can be said that the probability that the person has the wireless tag is high, and therefore the association level is increased. Similarly, the smaller the azimuth deviation amount or the higher the direction detection accuracy of the wireless tag, the higher the association level. Depending on the security level that is required, for example, a staying event of a specific behavior may be allowed to have a relatively low association level, while an admission event must be a high association level. Operation is possible.

10 wireless tag, 20 reader, 30 ranging sensor, 40 monitoring device, 42 storage unit, 44 reporting unit, 46 reporting unit, 48 monitoring processing unit, 50 communication line, 60 security center

Claims (4)

A moving body monitoring system for identifying whether a moving body is accompanied by a portable device,
Object detection means for detecting the position of the moving body in the monitoring area;
Receiving means for receiving a radio signal emitted by the portable device;
Azimuth detecting means for detecting the position or azimuth of the portable device from the received radio signal;
When the position of the mobile unit and the position of the portable device, or the direction specified by the position of the mobile unit and the direction of the portable device are compared, and the degree of coincidence of the position or the direction satisfies the first criterion Corresponding processing means for associating the mobile body with the portable device;
Tracking means for recording the position of the detected moving body together with the result of the association,
In the case where the correspondence processing means satisfies the second criterion that the degree of coincidence of the position or the direction with the portable device is looser than the first criterion for the mobile object associated with the portable device. Maintain the mapping,
A moving body monitoring system characterized by that.   The correspondence processing means associates a mobile object that is not associated with the portable device with the portable device when an amount of deviation of the position or the direction from the portable device is within a first reference value. For a mobile object associated with the portable device, if the amount of deviation from the portable device is within a second reference value larger than the first reference value, association with the portable device is performed. The moving body monitoring system according to claim 1, which is maintained.   The correspondence processing means associates the mobile body with the mobile device on the condition that the position or orientation of the mobile device changes following a change in position caused by the movement of the mobile body. The mobile body monitoring system according to claim 1 or 2. Furthermore, when the specific action by the mobile body is detected according to the history of the position of the mobile body, and the specific action by the mobile body is detected, the mobile device is not associated with the mobile body. The moving body monitoring system according to claim 1, further comprising an abnormality determining unit that determines an object.

Images