JP2008077361A - Monitoring method and monitoring system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To estimate location of a person in a given building. <P>SOLUTION: A monitoring system 100 includes detectors 10a to 10c represented by a detector 10, a head-count obtaining device 20, and a processor 30. Also, an entrance area 120 defined to overlap with a third area 110c is monitored by the head-count obtaining device 20 and head-count information obtained is output to the processor 30. The plurality of detectors 10 monitor respective assigned areas, detect infrared ray emitted from people or the like, and output them as movement information to the processor 30 at predetermined intervals. The processor 30 uses the movement information and the head-count information obtained to estimate presence of people or the like in each area. This estimation process may be executed by employing a method using the previous movement information or the like in the same area, a method using movement information or the like in other areas at the same time, and a method using future movement information or the like in the same area. <P>COPYRIGHT: (C)2008,JPO&INPIT

Description

  The present invention relates to a monitoring technique, and more particularly to a monitoring method for monitoring the arrangement of a person in a room such as a museum and a monitoring system using the same.

In recent years, monitoring systems have been introduced in various situations. The monitoring system is used, for example, as a part of a security system or as a system for assisting care. In general, in a system for monitoring a person, the person is monitored from the outside by using a monitoring camera or the like, and the movement of the person is monitored. Moreover, the state of the person is estimated by obtaining information about the person, for example, physical information such as voice and body temperature, from a terminal carried by the person to be monitored. In addition, the state of the subject is estimated using a sensor installed on the environment side, the state of the subject is acquired using a sensor that directly obtains information from the subject, and the information of the subject is combined and used. Some estimate the state (see, for example, Patent Document 1).
JP 2006-092356 A

  In general, an installed surveillance camera may be a psychological pressure for a person to be monitored. In particular, if a surveillance camera is installed in a museum or the like, the presence of the surveillance camera may be a concern, and viewers may not be able to concentrate on seeing art. Also, a great deal of effort is required to analyze the monitored image. In addition, the estimation of the state using a mobile terminal or the like has a problem of low estimation accuracy. In addition, the technique disclosed in Patent Document 1 has a problem that a viewer must have a terminal.

  This invention is made | formed in view of such a condition, The objective is to provide the monitoring method which can estimate the arrangement | positioning of the person in a predetermined hall easily, and the monitoring system using the same.

  In order to solve the above problems, a monitoring system according to an aspect of the present invention includes a plurality of detection devices that respectively detect movements of an object such as a human being or an animal, and a processing device that processes outputs of the plurality of detection devices. Each of the plurality of detection devices includes a detection unit that detects the presence or absence of movement of the object in the assigned region, and outputs detection information from the detection unit to the processing device at a predetermined output cycle. And an output unit. The processing device includes a receiving unit that receives detection information output from the plurality of detection devices, and an estimation unit that estimates the position of the object based on the detection information received by the receiving unit. The output periods at the output units of the respective detection devices may be synchronized.

  According to this aspect, the position of the person can be easily estimated by estimating the positions of the objects in the plurality of regions based on the detection information output from the plurality of detection devices.

  The detection device is an infrared sensor that detects a change in infrared with a sampling period shorter than the output period, and when the detection unit detects the movement of the object, it sets the state value of the detection information to ON, When the ON value is maintained and the output unit outputs the detection information, the state value may be set to OFF.

  According to this aspect, when the movement of the object is detected by the infrared sensor, the state value of the detection information is set to ON and the ON value is maintained, and when the output unit outputs the detection information, the state value is set to OFF. Thus, a state transition model having a predetermined rule is constructed, and by using the state transition model, it is possible to effectively estimate the arrangement of persons.

  The processing apparatus may further include a recording unit that records and holds, for each region, detection information that the receiving unit receives at a predetermined cycle. The estimation unit uses the detection information of a certain area recorded in the recording unit and the detection information of the same area continuously recorded in the detection information to estimate the presence / absence of an object in the area. Also good.

  According to this aspect, the presence / absence of an object in the area is estimated using the detection information of a certain area recorded in the recording unit and the detection information of the same area continuously recorded in the detection information. Thus, the arrangement of persons can be estimated efficiently.

  The estimation unit uses the detection information of a certain region recorded in the recording unit at a certain time and the detection information of the same region detected before that detection information, and the presence of the object at that time in that region is detected. You may perform the sequential determination process which estimates the presence or absence.

  Here, the “detection information of the same region detected before the detection information” includes detection information detected at an arbitrary past time when a certain time is used as a reference. According to this aspect, using the detection information of a certain area recorded in the recording unit at a certain time and the detection information of the same area detected before that detection information, the existence of the object at that time in that area By executing the sequential determination process for estimating the presence / absence of the person, the arrangement of the person can be easily estimated.

  The estimation unit uses the detection information at a certain time of a certain region recorded in the recording unit and the detection information of the same region detected after the detection information to determine whether the object exists at the time of that region. You may perform the reverse determination process to estimate.

  Here, the “detection information of the same area detected after the detection information” includes detection information detected at a future time when a certain time is used as a reference. According to this aspect, using the detection information at a certain time of a certain area recorded in the recording unit and the detection information of the same area detected after that detection information, the presence of the object at the predetermined time in that area By executing the reverse determination process for estimating the presence or absence of the person, the arrangement of the person can be easily estimated.

  The estimation unit uses the detection information at a certain time of a certain region recorded in the recording unit and the detection information of the same region detected before the detection information to determine whether the object exists at the time of that region. If the presence / absence of the target cannot be estimated even when the sequential determination process is performed, the detection information at the time of the area recorded in the recording unit is the same as that detected after the detection information. You may perform the reverse determination process which estimates the presence or absence of the target object in the time of an area | region using the detection information of an area | region.

  According to this aspect, if the presence / absence of the target object cannot be estimated even if the forward determination process is executed, the past and future information on the area that cannot be estimated by executing the reverse determination process. Can be used efficiently, and the arrangement of people can be estimated easily.

  The estimation unit uses the detection information at a certain time of a certain region recorded in the recording unit and the detection information of the same region detected after the detection information to determine whether the object exists at the time of that region. If the presence / absence of the object cannot be estimated even after executing the reverse determination process to be estimated, the detection information at the time of the area recorded in the recording unit is the same as that detected before the detection information. You may perform the sequential determination process which estimates the presence or absence of the target object in the time of the area | region using the detection information of an area | region.

  According to this aspect, if the presence / absence of the object cannot be estimated even when the reverse determination process is executed, the past and future information can be efficiently used by executing the forward determination process, It is possible to easily estimate the arrangement of the person in the area that could not be estimated.

  The processing apparatus may further include a recording unit that records and holds, for each region, detection information that the receiving unit receives at a predetermined cycle. The estimation unit estimates the presence / absence of an object in the region or the surrounding region by using the detection information recorded in the recording unit and the detection information recorded for the region around the region. May be.

  According to this aspect, the presence / absence of the object in the area or the surrounding area is estimated using the detection information recorded in the recording unit and the detection information recorded in the surrounding area. By doing so, the arrangement of persons can be estimated efficiently.

  The estimation unit executes the same determination process for estimating the presence or absence of an object at the same time around an arbitrary region, using detection information of a plurality of regions recorded at the same time in the recording unit. May be.

  According to this aspect, by using the detection information of a plurality of areas recorded at the same time on the recording unit, by executing the same determination process for estimating the presence or absence of the object at the same time around any area The person's arrangement can be easily estimated for the area that could not be estimated.

  The processing apparatus may further include a recording unit that records and holds, for each region, detection information that the receiving unit receives at a predetermined cycle. The estimation unit uses the detection information of a certain area recorded in the recording unit at a certain time and the detection information of the same area detected after that detection information, and the presence or absence of an object in that area at that time Using the detection information at a certain time of a certain region recorded in the recording unit and the detection information of the same region detected before the detection information, and the time of that region If the presence / absence of an object cannot be estimated even after executing the forward determination process for estimating the presence / absence of the object in the case, detection information of a plurality of areas recorded at the same time as the time in the recording unit May be used to execute the same determination process for estimating the presence / absence of an object at the time of an area around an arbitrary area.

  According to this aspect, if the presence / absence of the object cannot be estimated even after the reverse determination process or the forward determination process is performed, the region that cannot be estimated by executing the same determination process It is possible to easily estimate the arrangement of persons.

  The processing apparatus may further include a recording unit that records and holds, for each region, detection information that the receiving unit receives at a predetermined cycle. The estimation unit executes the same time determination process for estimating the presence / absence of an object at a time in a region around an arbitrary region, using detection information of a plurality of regions recorded at the same time in the recording unit. However, if the presence or absence of the target cannot be estimated, the detection information at the time of a certain area recorded in the recording unit and the detection information of the same area detected after the detection information are used to The detection process at the time of a certain region recorded in the recording unit and the detection information of the same region recorded before the detection information It may be used to execute the sequential determination process for estimating the presence / absence of an object at the time of the region.

  According to this aspect, if the presence / absence of the object cannot be estimated even if the same determination process is executed, the region that could not be estimated by executing the reverse determination process or the sequential determination process It is possible to easily estimate the arrangement of persons.

  The processing apparatus may further include a recording unit that records and holds, for each region, detection information that the receiving unit receives at a predetermined cycle. The estimation unit detects the detection information at the first time of a certain region recorded in the recording unit and the detection information before the detection information until the presence / absence of the object in all the regions at all times is estimated. Using the detection information of the same area, the sequential determination process for estimating the presence or absence of the object at the first time in the area, and the detection information of the plurality of areas recorded at the second time in the recording unit In addition, the same determination process for estimating the presence / absence of an object in an area around an arbitrary area, detection information at a third time of a certain area recorded in the recording unit, and detection after the detection information A plurality of determination processes may be sequentially executed out of the three determination processes including the reverse determination process for estimating the presence / absence of an object at the third time in the area using the detection information of the same area.

  The estimation unit may repeatedly execute the sequential determination process, the same determination process, the reverse determination process, and the same determination process in the order of arrangement.

  According to this aspect, among the three determination processes of the forward determination process, the same determination process, and the reverse determination process, by sequentially executing a plurality of determination processes, an area that could not be estimated can be simplified. A person's arrangement can be estimated.

  According to this aspect, it is possible to easily estimate the arrangement of a person in an area that cannot be estimated by repeatedly executing the sequential determination process, the same determination process, the reverse determination process, and the same determination process in order.

  When the estimation unit cannot estimate the presence or absence of the target, the output unit may set the output cycle to be short. The output unit may set the output cycle to be shorter as the number of objects in the plurality of regions increases. According to this aspect, by setting the output cycle to be short, it is easy to estimate the presence / absence of the object, and the arrangement of the person can be estimated easily. Moreover, the time required for the estimation process can be reduced by setting the output cycle to be shorter as the number of objects in the plurality of regions increases.

  The processing apparatus may notify the output unit of an output request signal for requesting output. The output unit may output detection information to the processing device in response to the notified output request signal. The output unit holds the detection information from the detection unit until the output request signal is notified. When the output request signal is notified from the processing device, the output unit outputs the held detection information to the processing device. May be. The output unit collectively outputs all detection information stored until the output request signal is notified to the processing device. In this case, since the processing apparatus can control the output of the output unit, the processing apparatus can be designed regardless of the specifications of the output unit, and the design flexibility can be improved.

  Another aspect of the present invention is a monitoring method. This method detects movements of objects such as humans or animals in a plurality of regions, stores each detected detection information in a predetermined cycle, and based on each stored detection information, The position of the object in the region is estimated.

  It should be noted that any combination of the above-described constituent elements and a conversion of the expression of the present invention between a method, an apparatus, a system, a recording medium, a computer program, etc. are also effective as an aspect of the present invention.

  According to the present invention, it is possible to easily estimate the arrangement of persons in a predetermined hall.

  Before the embodiments of the present invention are specifically described, an outline is first described. The embodiment of the present invention relates to a monitoring system for monitoring a human or an animal (hereinafter referred to as a person), and is suitable for monitoring a person or the like in a relatively uncongested premises such as a museum. The monitoring system according to the embodiment of the present invention has a function of estimating the arrangement of persons or the like at an arbitrary time point using a plurality of infrared detection devices.

  The plurality of infrared detection devices are arranged on a ceiling or the like so that all locations in the hall (hereinafter also referred to as “regions”) can be comprehensively monitored. Each infrared detecting device detects a change of an infrared ray emitted from a person or the like in a predetermined area of a specific premises, thereby detecting a movement of the person or the like in the area. Here, when an action of a person or the like is detected, a state value indicating the action is set to ON and the ON value is maintained. In addition, the infrared detection device periodically outputs operation information indicating the movement of a person or the like detected in each assigned area to the processing device, and sets the state value to OFF in response to the output.

  As a result, a state transition model that associates the motion information with the presence / absence of a person is configured. The processing device stores the operation information output from the detection device corresponding to each area. The processing device estimates the presence / absence of a person at an arbitrary time and in an arbitrary region by using a state transition model in which the motion information and the presence / absence of a person are related, using a plurality of stored movement information. . Thereby, arrangement | positioning of the person etc. in arbitrary time can be estimated.

  The estimated arrangement of persons or the like can be used as information for guiding a disabled person or the like, for example. In this case, it is possible to guide more safely while selecting a non-congested place by periodically transmitting arrangement information of persons or the like existing in the hall to a guide or a guidance robot. In addition to the monitoring system of the present embodiment, security can be further improved by installing a monitoring camera at the entrance / exit of the premises. The position of the person confirmed by the monitoring camera is estimated at an arbitrary time by the monitoring system of this embodiment, and the locus of the person or the like can be captured from the position. This technique can also be applied, for example, to identify the movement of thieves in exhibitions such as jewelry. Compared to using multiple surveillance cameras to cover all the venues of the exhibition, it is possible to realize a security system that is almost as advanced.

  FIG. 1 is a diagram illustrating a configuration example of a monitoring system 100 according to an embodiment of the present invention. The monitoring system 100 includes detection devices 10 a to 10 c typified by the detection device 10, a number acquisition device 20, and a processing device 30. In FIG. 1, the first area 110a is assigned to the detection apparatus 10a, the second area 110b is assigned to the detection apparatus 10b, and the third area 110c is assigned to the detection apparatus 10c. . Each detection device 10 can monitor the entire area such as a museum, for example, by monitoring the area assigned to each detection apparatus 10. In addition, the entrance / exit area 120 defined to overlap the third area 110c is monitored by the number-of-people acquisition apparatus 20. For convenience of explanation, three regions and one entrance / exit region are shown, but a number other than 3 regions may exist, or two or more entrance / exit regions may exist. In addition, each area is wide enough to accommodate only one person.

  The plurality of detection devices 10 monitor the assigned areas, and detect infrared rays emitted from a person or the like. The detection device 10 analyzes the detected infrared rays and generates movement information of a person or the like. Further, the detection device 10 outputs the generated motion information to the processing device 30 at a predetermined cycle. Further, the detection device 10 initializes motion information in accordance with an instruction from the processing device 30. The detection apparatus 10 may initialize the motion information in response to its own output.

  FIG. 2 is a diagram illustrating a configuration example of the detection device 10 of FIG. The detection device 10 includes an infrared detection unit 12, an output unit 16, and an initialization unit 18. For example, the infrared detector 12 detects infrared rays emitted from a person or the like in the allocated area by an infrared sensor or the like. A Fresnel lens or the like may be attached to the infrared sensor. As a result, the target location to be detected can be narrowed down, and the movement of a person or the like in a specific narrow range can be detected. In addition, a dual element type may be used as an infrared sensor to reduce detection of background fluctuations. Further, by connecting a bandpass filter to an amplifier for amplifying the detected infrared signal, it is not necessary to detect a slow change or a too fast change. Further, it is not necessary to use a window comparator to detect a motion with little change.

  The infrared detection unit 12 notifies the output unit 16 of the detected infrared change. Here, the infrared detection unit 12 moves when a person or the like moves from another region to the region assigned to the detection apparatus 10 or when the person moves from the assigned region to another region. Information indicating the operation is notified to the output unit 16 as information. In addition, even when a person or the like moves in the allocated area, information indicating the operation is notified to the output unit 16. On the other hand, when a person or the like does not move within the allocated area, or when no one has moved within the allocated area, information indicating no operation is output to the output unit 16 as movement information. Notice. In the following, for the convenience of explanation, information indicating the operation is indicated by 1, and information indicating the non-operation is indicated by 0.

  The output unit 16 transmits the notified information to the processing device 30 at a predetermined cycle. When the information indicating the operation is notified in the middle of the predetermined cycle, the information indicating the operation is transmitted to the processing device 30 at the head of the next cycle. The transmission cycle may be set according to the moving speed of a person or the like, the number of people in a plurality of areas, or may be a fixed value, for example. Note that the transmission cycle in the output unit 16 is longer than the detection cycle in the infrared detection unit 12.

  The initialization unit 18 initializes the output of the infrared detection unit 12 based on an instruction from the processing device 30. The initialization includes setting information to be output to the output unit 16 or the processing device 30 (hereinafter referred to as output information) to information indicating no operation. Note that the detection device 10 may not include the initialization unit 18. In this case, the infrared detection unit 12 may initialize the output information in response to the output from the output unit 16 to the processing device 30.

  FIG. 3 is a timing chart showing an input / output example of the infrared detection unit 12 of FIG. The horizontal axis indicates time. The vertical axis indicates motion information. The infrared detector 12 detects a change in infrared rays with a period of 0.5. In addition, the infrared detection unit 12 outputs output information to the processing device 30 in one cycle. In addition, it is assumed that at each timing of t = 3 and 10, it has moved from another area into the area assigned to the detection device 10. In addition, it is assumed that the region has been moved from the region assigned to the detection device 10 to another region at the timing of t = 5. In addition, it is assumed that a person or the like moves in the region at the timing of t = 13.

  First, at t = 0 to 3, since there is no person in the area assigned to the detection apparatus 10 and the person does not move from another area, the infrared detection unit 12 transmits infrared rays. Information indicating no operation is not notified to the output unit 16. Next, at t = 3, since it has moved from another region into the region assigned to the detection device 10, the infrared detection unit 12 detects a change in infrared and notifies the output unit 16 of the change. At t = 4, the output unit 16 transmits information indicating the operation to the processing device 30, and the infrared detection unit 12 initializes the motion information.

  At t = 5, because the region has been moved from the region assigned to the detection device 10 to another region, the infrared detection unit 12 detects a change in the infrared and notifies the output unit 16 of information indicating the operation as motion information. The infrared detection unit 12 continues to output information indicating the operation until t = 8, which is the next initialization timing. At t = 8, the output unit 16 transmits information indicating an operation to the processing device 30 as motion information, and the infrared detection unit 12 initializes the output. At t = 8-10, there is no person in the area assigned to the detection apparatus 10 and no person moves from another area, so the infrared detection unit 12 cannot detect infrared rays. First, information indicating no operation is notified to the output unit 16.

  At t = 10, the infrared detection unit 12 detects the infrared ray and notifies the output unit 16 because it has moved from another region into the region assigned to the detection device 10. The infrared detection unit 12 continues to output the information indicating the operation until t = 12, which is the next initialization timing. At t = 12, the output unit 16 transmits information indicating the operation to the processing device 30. Here, although the infrared detection unit 12 is initialized once, a person or the like has moved within the area assigned to the detection device 10. Therefore, the infrared detection unit 12 detects infrared rays and outputs information indicating the operation. Notify The infrared detecting unit 12 continues to output the information indicating the operation until t = 16 which is the next initialization timing. At t = 16, the output unit 16 transmits information indicating the operation to the processing device 30.

  In summary, the output unit 16 outputs information indicating non-operation to the processing device 30 at the end of the cycle when a change in infrared rays has never been detected within the predetermined cycle. Further, when a change in infrared rays is detected even once within a predetermined cycle, the output unit 16 outputs information indicating the operation to the processing device 30 at the end timing of the cycle.

  Returning to FIG. The number acquisition device 20 detects the number of persons and the like existing in the first area 110a to the third area 110c by monitoring the entrance / exit area 120. The number-of-people acquisition device 20 may radiate two infrared lasers in the horizontal direction at the entrance / exit, for example. The two infrared lasers may be provided so as to have a predetermined distance from the exit side of the entrance to the entrance side. The predetermined interval may be adjusted according to the speed at which a person or the like walks. Thus, when a person or the like shuts off the two infrared lasers in order, it is possible to determine whether to enter or leave the room according to the order of shutting down. Here, the number of persons and the like existing in the first area 110a to the third area 110c can be derived by subtracting the number of persons who have left the room from the number of persons who have entered the room. If there are a plurality of entrances / exits in the area, the number of persons, etc. in the area can be derived in the same manner by providing the number-of-people acquisition device 20 for each entrance / exit and using the total number of entrances and exits. The derived number of persons (hereinafter referred to as number information) is notified to the processing device 30 at a predetermined cycle. The predetermined cycle may be the same as the cycle output from the detection device 10 to the processing device 30 or may be a cycle of an integer. Moreover, the period which each detection apparatus 10 outputs to the processing apparatus 30 shall be synchronized.

  The processing device 30 acquires motion information from the detection device 10 at a predetermined cycle. For example, when the output shown in FIG. 3 is acquired, 1 is acquired at timings of t = 4, 8, 12, and 16, respectively. Further, the processing device 30 acquires and stores the number information from the number acquisition device 20. Further, each time the motion information is acquired, the processing device 30 uses the acquired motion information and the number of people information to estimate the presence / absence of a person or the like in each area (hereinafter referred to as a person flag). The estimation process is executed for each region every time motion information or the like is acquired. Although details will be described later, the estimation processing is performed using a method using past motion information or the like in the same region (hereinafter referred to as a forward determination process) and a method using motion information or the like of another region at the same time (hereinafter referred to as the same time) This is implemented by using any one of a method using determination information), a method using future motion information in the same region (hereinafter referred to as reverse determination processing), or a method combining them. The estimated person flag is stored in the memory. If these methods cannot be used for estimation, the person flag is stored as “indeterminate”.

  FIG. 4 is a diagram illustrating a configuration example of the processing device 30 of FIG. The processing device 30 includes a first receiving unit 32, a first memory 34, a second receiving unit 36, a position estimating unit 38, a second memory 40, a control unit 42, and an instruction unit 44. The first receiving unit 32 receives the motion information output from the detection device 10 and stores it in the first memory 34 while associating it with the received time or the detected time. The second receiving unit 36 receives the number information transmitted from the number acquisition device 20 and stores it in the first memory 34 while associating it with the received time. The position estimation unit 38 uses the motion information and the number of people information stored in the first memory 34 to estimate the presence / absence of a person or the like for each time or for each region, and stores it in the second memory 40. To do. The control unit 42 controls the position estimation unit 38. In addition, the control unit 42 controls the instruction unit 44 to instruct the detection device 10 to perform initialization. The instruction unit 44 instructs the detection apparatus 10 to initialize.

  These configurations described above can be realized in hardware by a CPU, memory, or other LSI of an arbitrary computer, and in software, it can be realized by a program having an optimization function loaded in the memory. Here, the functional blocks realized by the cooperation are depicted. Accordingly, those skilled in the art will understand that these functional blocks can be realized in various forms by hardware only, software only, or a combination thereof.

Here, the relationship between the motion information and the number of people information stored in the first memory 34 will be described using a state transition diagram. FIG. 5 is a state transition diagram 200 showing an example of state transition in the monitoring system 100 of FIG. The state transition diagram 200 includes states S0 to S3 and shows states in an arbitrary region. Each state is defined by the motion information and the number of people information in the area as follows. In addition, when a person etc. does not exist, it is shown by person flag = 0, and when a person etc. exist, it is shown by person flag = 1.
State S0: motion information = 0, person flag = 0
State S1: motion information = 1, person flag = 1
State S2: motion information = 0, person flag = 1
State S3: motion information = 1, person flag = 0

  The state S0 indicates that the latest motion information is 0 and no person or the like exists in the area. Similarly, the state S1 indicates that the latest motion information is 1, and that a person or the like exists in the area. The state S2 indicates that the latest motion information is 0 and that a person or the like exists in the area. The state S3 indicates that the latest motion information is 1 and that no person or the like exists in the area.

  Each state transitions to another state according to the motion information at the next time. The time indicates the time that the output unit 16 outputs to the processing device 30. Further, the next time indicates the time to be output next after the output unit 16 outputs to the processing device 30. This will be specifically described. In the state S0, when the motion information at the next time is 0, for example, when there is no movement of a person or the like from the surrounding area and a state in which no person or the like continues exists, the state is changed by the first transition 202. The state of S0 is maintained. On the other hand, when the motion information at the next time is 1, for example, when a person or the like has moved from the surrounding area, the state transitions to the state S1 by the second transition 204.

  In the state S1, when the motion information at the next time is 0, for example, when a person or the like existing in the region is stationary in the region, the state transitions to the state S2 by the third transition 206. On the other hand, when the motion information at the next time is 1, two transitions may occur. For example, when a person or the like moves within the area but a person or the like still exists, the state S1 is maintained by the fourth transition 208. When a person or the like existing in the area moves to the surrounding area at the next time, the state transitions to the state S3 by the fifth transition 210. Here, the information that can be acquired by the processing device 30 is only the motion information and the number of people information, and the specific information such as “when a person or the like moves in the area but a person or the like still exists” is used. Since it is not information, if the motion information is 1 in the state of state S1, the transition destination cannot be specified. Hereinafter, such a state is referred to as “indeterminate”. Even once, if the state becomes uncertain, the state in the same area after that time also remains uncertain. A method for removing the indeterminate state will be described later.

  In the state S2, when the motion information at the next time is 0, for example, when a person or the like existing in the area is stationary in the area, the state S2 is continued by the sixth transition 212. On the other hand, when the motion information at the next time is 1, two transitions may occur as in the state S1 described above. For example, when a person or the like moves in the region but a person or the like still exists, the state transitions to the state S1 by the seventh transition 214. When a person or the like existing in the area moves to the surrounding area at the next time, the state transits to the state S3 by the eighth transition 216. Therefore, similarly to the state S1, even in the state S2, if the motion information is 1, the transition destination cannot be specified.

  In the state S3, when the motion information at the next time is 0, for example, when there is no movement of a person or the like from the surrounding area and a state in which no person or the like continues continues, Transition to S0. On the other hand, when the motion information at the next time is 1, for example, when a person or the like has moved from the surrounding area, the state transitions to the state S1 by the ninth transition 218.

The above is summarized as follows.
In the state S1 and the state S2, when the motion information at the next time is 1, the state after the next time and the person flag are uncertain.
○ In cases other than the above, the state and the person flag are fixed.

  Here, the operation | movement concerning the estimation process of the processing apparatus 30 is demonstrated. As described above, the processing device 30 estimates the presence / absence of a person or the like at an arbitrary time using the motion information and the number of persons information acquired for each region. Specifically, the presence / absence of a person or the like is estimated using any one of the determination processes (1) to (3) described below. Here, the methods shown in (2) and (3) are mainly used when an uncertain state occurs. In the following, for convenience of explanation, the next time is expressed as “present”. Also, the time immediately before the next time is expressed as “past”.

(1) Sequential Determination Processing The transitions in the state transition diagram 200 in FIG. 5 are summarized as shown in FIG. FIG. 6 is a diagram showing a state relationship diagram 400 between the current state and the past state in the state transition diagram 200 of FIG. The state relationship diagram 400 includes a past state column 410, a current state column 420, and a transition column 430. Here, the past state refers to a state at the previous time with respect to the current time (hereinafter also referred to as “reference time”). The state at the previous time indicates, for example, the motion information detected at the time immediately before the motion information detected at the reference time. S0 to S3 shown in the past state column 410 or the current state column 420 correspond to the states S0 to S3 in the state transition diagram 200 of FIG. In addition, among X / Y in parentheses shown on the right side of S0 to S3, X indicates motion information, and Y indicates a person flag. The transition column 430 indicates a transition number when transitioning from a past state to the current state. 202 to 220 shown as transition numbers respectively correspond to the first transition 202 to the tenth transition 220 in the state transition diagram 200 of FIG. In addition, “x” indicates that there is no transition.

  Here, a method for estimating the person flag in the current state using the past state will be described. As illustrated, when the past state is S0 or S3, the current state transitions to either S0 or S1. Here, the person flags in the past S0 and S3 are both 0. Therefore, when the person flag in the past state is 0, the current state is changed to S0 or S1. Further, S0 and S1 in the current state can be identified by motion information in the past or present state. In summary, when the person flag in the past state is 0, the person flag in the current state can be determined based on the motion information of the current state.

  Similarly, when the past state is S1 or S2, the current state transitions to either S1, S2, or S3. Here, the person flags in the past states S1 and S2 are all 1. Therefore, a person flag of 1 in the past state transitions to S1, S2, or S3 of the current state. Also, among the current states S1, S2 and S3, the motion information of S2 in the current state is 0, while S1 and S3 are 1, so if the motion information is 0, the current state is Can be identified. In summary, if the person flag in the past state is 1 and the motion information in the current state is 0, the person flag in the current state can be determined. On the other hand, if the person flag in the past state is 1 and the current motion information is 1, the current state is not fixed and becomes uncertain.

  FIG. 7 is a flowchart illustrating an operation example of the sequential determination process according to the processing device 30 of FIG. This process may be executed each time motion information for each region is acquired. Moreover, it may be executed when the motion information of a specific area is acquired. First, the position estimation unit 38 reads the motion information at the reference time stored in the first memory 34 and the previous person flag stored in the second memory 40 (S10). If the previous person flag is 0 (Y in S12), the process proceeds to the determination process in S14, and if 1 (N in S12), the process proceeds to the determination process in S20. In S14, when the read motion information is 0 (Y in S14), it is determined that the person flag at the reference time is 0. Here, the position estimation unit 38 sets the person flag at the reference time in the second memory 40 to 0 (S16). On the other hand, if the read motion information is 1 (N in S14), it is determined that the person flag at the reference time is 1. Here, the position estimation unit 38 sets the person flag at the reference time in the second memory 40 to 1 (S18).

  In S20, if the previous person flag is 1 (Y in S20), the process proceeds to S22. If the previous person flag is not 1 (N in S20), the process proceeds to S24. In S24, when the read motion information is 0 (Y in S22), it is determined that the person flag at the reference time is 1. Here, the position estimation unit 38 sets the person flag at the reference time in the second memory 40 to 1 (S18). On the other hand, when the read motion information is 1 (N in S22), it is determined that the person flag at the reference time is indeterminate. Here, the position estimation unit 38 sets information indicating uncertainness as the person flag at the reference time in the second memory 40 (S24).

(2) Same time determination process This process is used when there is a region where the person flag is indeterminate in the current or past state. This same-time determination process is a method of determining a person flag in an area in which the person flag is not fixed at the reference time, using the number information and the number of areas in which the person flag is fixed at the reference time. Here, among the other areas in which the person flag is fixed at the same time as the reference time, if the number of areas in which the person flag is 1 is equal to the number information, all the areas that are indeterminate The person flag can be determined to be 0. In addition, when the value obtained by subtracting the number of areas in which the person flag at the same time is 1 from the number information is equal to the number of indeterminate areas, it is determined that the person flags in all indeterminate areas are 1. it can. In other cases, the indeterminate state cannot be removed by the same determination process.

  FIG. 8 is a flowchart showing an operation example of the same time determination processing according to the processing device 30 of FIG. If there is an area where the person flag is uncertain at an arbitrary time, this process may be executed for the area at that time. First, the position estimating unit 38 compares the number of people information stored in the first memory 34 with the number of areas in which the person flag at the time stored in the second memory 40 is set to 1 (S26). . Here, when the number information and the number of areas are equal (Y of S26), the person flag is set for all areas where the person flag at the time stored in the second memory 40 is stored as indefinite. It is set to 0 (S28). When the number information is different from the number of areas (N in S26), and the value obtained by subtracting the number of areas from the number information is equal to the number of areas where the person flag is set as indeterminate (Y in S30), the person flag is set to 1 for all areas where the person flag at the time stored in the second memory 40 is stored as indeterminate (S32). On the other hand, when the number information and the number of areas are different (N in S26), the value obtained by subtracting the number of areas from the number information is the number of areas where the person flag is set as indeterminate. If they are not equal (N in S30), the process is terminated.

(3) Reverse determination process This process is performed when a person flag in an area is uncertain in the past state, and then the person flag in the area is determined by the same determination process shown in (2). Used for. This reverse time determination process is a method of using the person flag at the reference time to determine the person flag in the past that has been indeterminate from the past. Here, the reverse determination process will be described with reference to FIG.

  As shown in FIG. 6, when the current state is S0, it can be said that the state in the past has transitioned from either S0 or S3. Here, the person flags in the past S0 and S3 are both 0. Therefore, when the current state is S0, the person flag in the past state is 0 regardless of the motion information. Similarly, it can be said that the current state has transitioned to S2 or S3 from either the previous state S1 or S2. Here, the person flags in the past S1 and S2 to be confirmed are both 1. Therefore, when the current state is S2 or S3, the person flag in the past state is 1 regardless of the motion information. On the other hand, it can be said that the current state S1 has transitioned from any of the past states S0 to S3. Since these transitions are not discriminative, when the current state is S1, the past state is also uncertain, and the indeterminate cannot be removed in the reverse determination process.

  FIG. 9 is a flowchart illustrating an operation example of the reverse determination processing according to the processing device 30 of FIG. This process may be executed in response to the confirmation of the current person flag for a specific area, for example, an area where the person flag has been uncertain in the past. First, the position estimation unit 38 reads out the current motion information stored in the first memory 34 and the current person flag stored in the second memory 40 (S34). Here, if the current person flag is 0 (Y in S36), the process proceeds to the determination process in S38, and if 1 (N in S36), the process proceeds to the determination process in S44. In S38, when the read motion information is 0 (Y in S38), it is determined that the current person flag is 0. Here, the position estimation unit 38 sets the past person flag in the second memory 40 to 0 (S40). On the other hand, if the read motion information is 1 (N in S38), it is determined that the current person flag is 1. Here, the position estimation unit 38 sets the past person flag in the second memory 40 to 1 (S42).

  In S44, if the current person flag is 1 (Y in S44), the process proceeds to S46, and if the current person flag is not 1 (N in S44), the process proceeds to S48. In S46, when the read motion information is 0 (Y in S46), it is determined that the past person flag is 1. Here, the position estimation unit 38 sets the current person flag in the second memory 40 to 1 (S42). On the other hand, if the read motion information is 1 (N in S46), it is determined that the current person flag is indeterminate. Here, the position estimation unit 38 sets information indicating uncertainness as the current person flag in the second memory 40 (S48).

(4) Combination of (1) to (3) There is a case where uncertainty cannot be eliminated by the method shown in any of (1) to (3). However, when the person flag in all the areas is fixed at any time, the probability of uncertainness can be improved by sequentially executing the methods (1) to (3).

  In order to improve the probability of escaping from uncertainty, it is necessary to determine people flag in all areas at any time after the person flag in any area at any time becomes uncertain. . It is preferable to use the method (2) in order to determine human flags in all areas at an arbitrary time. For example, when all persons on the premises where the monitoring system 100 according to the present embodiment is installed move to an area where the person flag is fixed, the person flag at that time is set to 1 by the method (1). And confirm. In this case, even if the person flag in the other area at the time is not fixed, it is determined by the method (2) that no person or the like exists in the other area. The person flag in the area is determined.

  In such a case, by using the confirmed time as the current state and further using the reverse determination process shown in (3), it is possible to escape from the indeterminate date. Even if all uncertainties in the past could not be removed, there is a case where uncertainties can be removed by using the method (2) again in the past time when the uncertainties could be removed. . Even in this case, when there is still a region where the uncertainty cannot be eliminated, the uncertainty can be further reduced by executing (1) or (3).

  By repeating the above process as much as possible, the uncertain area can be reduced as much as possible. Note that the uncertain state can also be removed by shortening the detection cycle in the detection apparatus 10. By shortening the detection cycle, the stable state of S0 or S3 continues in most areas, so the time at which the person flag in all areas is determined increases. Then, as described above, uncertain portions can be reduced by combining the methods (1) to (3). In addition, the monitoring system 100 according to the present embodiment is installed in a premises that is not relatively crowded, such as a museum. In such a case, it can be said that the persons on the premises often remain stationary. Then, since the probability of shifting to an indeterminate state is low, it can be said that the probability of escaping from indeterminacy is high by any one of the methods (1) to (3). In a crowded premises, it goes without saying that the same effect can be obtained if the detection cycle of the detection apparatus 10 is shortened according to the degree of congestion. Further, by setting the detection cycle short, the time required for the estimation process can be reduced.

  FIG. 10 is a flowchart showing an operation example of the processing apparatus 30 of FIG. First, the processing device 30 acquires motion information from each of the plurality of detection devices 10, and acquires the number of people information from the number of people acquisition device 20 (S50). Next, the processing device 30 stores the acquired motion information and the number of people information in the memory while associating with the current time (S52). Here, the processing device 30 executes the sequential determination process shown in FIG. 7 for each region in order to determine whether or not a person or the like exists in each region (S54). In response to the determination of the person flag indicating the presence / absence of a person or the like for all areas at the time, the processing device 30 determines whether there is an indeterminate area among the determined person flags. It is confirmed whether or not (S56). If there is no indeterminate area (N in S56), the process proceeds to S58. In S58, an end determination is executed (S58). In the end determination, the process ends according to an instruction from a person who operates the monitoring system 100 or when a preset time is reached (Y in S58). If not finished (N in S58), the process returns to S50.

  If there is an indeterminate area (Y in S56), the same determination process shown in FIG. 8 is executed (S60). If there is no region where the person flag is indefinite at the current time (N in S62), the process returns to S50. On the other hand, if there is an area where the person flag is uncertain (Y in S62), the reverse determination process (S64), the same determination process (S66), and the sequential determination process (S68) shown in FIG. Execute. The order of executing S64, S66, and S68 may be changed as appropriate. Further, the reverse determination process in S64 and the forward determination process in S68 are executed for an area where the person flag is indefinite. In addition, the same time determination process of S66 is executed for a time including an area where the person flag is indefinite. S64, S66, and S68 are repeated until there is no area where the person flag is indeterminate, or is repeated a preset number of times (S70). When there is no area where the person flag is indeterminate, or when the process is repeated a preset number of times (Y in S70), the process returns to S50.

  This will be described using a specific example. Here, an example in which the sequential determination process and the same determination process are used will be described. FIG. 11A is a diagram illustrating a region in the first operation example of the embodiment of the present invention. As shown in the figure, the premises where the monitoring system 100 is installed includes 16 areas A to P, and entrances are provided in the areas A and M. Each area is monitored by the corresponding detection device 10, and the movement of a person or the like in the area is detected. Further, the entrances and exits provided in the area A and the area M are monitored by the number-of-people acquisition device 20 to acquire the number of people on the premises.

  FIG.11 (b) is a figure which shows the example of the motion information concerning each area | region in the 1st operation example of embodiment of this invention. FIG. 11B shows the motion information stored in the first memory 34. It is assumed that these pieces of motion information are acquired between time t0 and time t7 for each of the areas A to P shown in FIG. As illustrated, motion information is stored as 0 or 1 for each time and each region. FIG.11 (c) is a figure which shows the example of a transition of the number of persons in the 1st operation example of embodiment of this invention. FIG. 11C shows the number information stored in the first memory 34. These pieces of number information are changes in the number of people on the premises from time t0 to time t7, and are acquired by the number acquisition device 20 that monitors the areas A and M shown in FIG. As shown, the number information is stored for each time.

  The flow of estimation processing by the processing device 30 will be described under the conditions shown in FIGS. FIG. 12A is a diagram illustrating an estimation result of the arrangement of each region in the first operation example of the embodiment of the present invention. The estimation result shown in FIG. 12A indicates the person flag stored in the second memory 40. In addition, the estimation result shown in FIG. 12A shows 0 or 1 only for the region where the person flag is confirmed, and “unknown” is illustrated as information indicating uncertainty in the region where the person flag is uncertain. Hereinafter, it demonstrates for every time.

  First, at time t0, since it is in an initial state, the person flag in all areas is estimated to be zero. Next, at time t1, the motion information shown in FIG. 11A is 1 only for the region A and 0 for the other regions. Further, the person flag at time t0 immediately before time t1 is 0 for all areas. Therefore, according to the sequential determination process shown in FIG. 7, the person flag is determined to be 1 for the area A, and the person flag is determined to be 0 for the other areas.

  Next, at time t2, the motion information shown in FIG. 11A is 1 only for the area A and the area B, and 0 for the other areas. The person flag at time t1 immediately before time t2 is 1 for the area A and 0 for all other areas. Accordingly, in the sequential determination process shown in FIG. 7, the person flag is determined to be indeterminate for the area A, the person flag is determined to be 1 for the area B, and the person flag is determined to be 0 for the other areas. Is done. Then, since there is an indeterminate region, the processing device 30 executes the same determination process. Here, the number of persons at time t2 shown in FIG. 11C is 1, and in FIG. 12A, the number of areas where the person flag is 1 at time t2 is 1. Therefore, the person flag in the area A at time t2 is determined to be 0.

  Thereafter, from time t3 to t7, the person flag in each area is determined in the same manner as at times t1 and t2. In this example, as the time advances, the number of areas in which the person flag is uncertain increases. In addition, from time t3 to t7, the number of persons at each time shown in FIG. 11C is equal to the number of areas in which the person flag is 1 at the corresponding time in FIG. Accordingly, in FIG. 12A, all the person flags in the area indicated as “not” are determined to be 0 by the same determination process.

  FIG. 12B is a diagram illustrating a trajectory of a person or the like estimated from the arrangement of each region in FIG. When there is no longer any area indicated as “not” for all areas at all times, the trajectory of a person or the like may be estimated. FIG. 12B shows that the persons X and Y have entered the respective rooms and moved to the respective areas at the timing indicated by the parentheses. The locus of the person X is indicated by a solid arrow, and the locus of the person Y is indicated by a broken arrow. As shown in the figure, the person X enters the area A at the timing of the time t1, and thereafter moves from the area A → the area B → the area C → the area D → the area H → the area L → the area P from the time t2 to the time t7. Indicates that In addition, the person Y enters the area M at the timing of time t4 and thereafter moves from the area M to the area I to the area J to the area F from time t5 to time t7.

  Furthermore, a specific example is shown. Here, an example in which sequential determination processing, same-time determination processing, and reverse determination processing are used will be described. FIG. 13A shows a region in the second operation example of the embodiment of the present invention. As shown in the figure, the premises where the monitoring system 100 is installed includes nine areas of areas A to I, and an entrance is provided in the area A. Each area is monitored by the corresponding detection device 10, and the movement of a person or the like in the area is detected. Further, the entrance / exit provided in the area A is monitored by the number-of-people acquisition device 20, and the number of people on the premises is acquired.

  FIG. 13B is a diagram illustrating an example of motion information regarding each region in the first operation example of the embodiment of the present invention. FIG. 13B shows motion information stored in the first memory 34. These pieces of motion information are assumed to have been acquired between time t0 and time t7 for each of the areas A to I shown in FIG. As illustrated, motion information is stored as 0 or 1 for each time and region. FIG. 13C is a diagram illustrating a transition example of the number of persons in the second operation example of the embodiment of the present invention. FIG. 13C shows the number information stored in the first memory 34. These pieces of number information are changes in the number of people on the premises between the times t0 and t7, and are acquired by the number acquisition device 20 that monitors the area A shown in FIG. As shown, the number information is stored for each time.

  The flow of estimation processing by the processing device 30 will be described under the conditions shown in FIGS. FIG. 14A is a diagram illustrating a first estimation result of the arrangement of each region in the second operation example of the embodiment of the present invention. The first estimation result indicates the estimation process at times t0 to t2, and indicates the person flag stored in the second memory 40. In addition, the estimation result shown in FIG. 14A shows 0 or 1 only for the region where the person flag is confirmed, and “unknown” is illustrated as information indicating the uncertainty in the region where the person flag is uncertain. Hereinafter, it demonstrates for every time.

  First, at time t0, since it is in an initial state, the person flag in all areas is estimated to be zero. Next, at time t1, the motion information shown in FIG. 13A is 1 only for the region A and 0 for the other regions. Further, the person flag at time t0 immediately before time t1 is 0 for all areas. Therefore, according to the sequential determination process shown in FIG. 7, the person flag is determined to be 1 for the area A, and the person flag is determined to be 0 for the other areas.

  At time t2, the motion information shown in FIG. 13A is 1 only for the regions A and B, and 0 for the other regions. The person flag at time t1 immediately before time t2 is 1 for the area A and 0 for all other areas. Accordingly, in the sequential determination process shown in FIG. 7, the person flag is determined to be indeterminate for the area A, the person flag is determined to be 1 for the area B, and the person flag is determined to be 0 for the other areas. Is done.

  FIG. 14B is a diagram illustrating a second estimation result of the arrangement of the areas in the second operation example of the embodiment of the present invention. The second estimation result indicates an estimation process at times t2 to t7. A region determined to be indeterminate by the forward determination processing is indicated by a broken-line circle. Further, an arrow that straddles two areas indicated by a broken-line circle is a sequential determination process, a simultaneous determination process, or a reverse determination process based on a person flag or the like of the area that starts from the arrow. Indicates that the person flag in the area at the end of the arrow is determined. The same applies to FIG. 14C described later.

  At time t2, since the area A exists as an indeterminate area, the processing device 30 executes the same determination process. Then, the number of people at time t2 shown in FIG. 13C is 1, and the number of areas where the person flag is 1 at time t2 is 1. Therefore, the person flag in the area A at time t2 is determined to be 0.

  Next, at time t3, the motion information shown in FIG. 13A is 1 only for the regions B and C, and 0 for the other regions. Further, the person flag at time t2 immediately before time t3 is 0 for all areas except for area B. Therefore, by the sequential determination process shown in FIG. 7, it is determined that the person flag is indeterminate for the area B, the person flag is determined to be 1 for the area C, and the person flag is 0 for the other areas. It is determined. Here, in the region B that has become indeterminate, the person flag is determined to be 0 by the same determination process.

  Similarly, at time t4, the person flag is determined except for the area C by the sequential determination process. In addition, in the region C that has become uncertain, the person flag is determined to be 1 by the same determination process. Further, at times t5 to t6, the sequential determination process and the same determination process are executed in the same manner. As shown in the figure, at times t5 to t6, even if the same determination process is executed, there is an area where the indeterminacy cannot be removed. At the last time t7, although there is an indefinite area by the sequential determination process, the person flag of all areas is determined by the same determination process.

  If the indeterminate state cannot be completely removed even if the same determination process is used, there are a plurality of regions in the indeterminate state. For example, at time t5, the area A and the area C are in an indeterminate state. As described above, when there is a region that cannot be completely indeterminate even if the sequential determination process and the same determination process are used, the region is often uncertain after the next time. For example, at times t6 and t7, the areas A and C continue to be indeterminate. On the other hand, even if there are a plurality of regions that are in an indeterminate state, there is an exception that indeterminateness can be completely removed. For example, at time t7, although region A, region B, region C, and region F are in an indeterminate state depending on the sequential determination process, the indeterminacy can be removed by the same determination process. Then, the person flag is fixed at time t7, while an uncertain region exists at t6 and t5 in the past from time t7. In such a case, by executing the reverse sequential determination process with reference to the time t7 when the person flag for all the regions is fixed, the uncertain state at the time t6 or t5, which is the past time, is removed. May be possible.

In summary, the following rules can be derived.
○ Even if sequential determination processing or simultaneous determination processing is used, if an area that cannot be undetermined has occurred, at a future time, until the person flag for all the areas is determined by the simultaneous determination processing, The indeterminate state continues.
The continuation of this indeterminate state is stopped when the person flag for all the areas is determined by the same determination process at a future time.
In addition, by executing the reverse determination process with reference to the future time when the person flag for all the areas is fixed, it is possible to escape from the uncertain state going back to the past.

  FIG. 14C is a diagram illustrating a third estimation result of the arrangement of the regions in the second operation example of the embodiment of the present invention. Here, it is shown that the indeterminate state can be escaped from the past by executing the reverse determination process based on the time t7 when the person flag for all the areas is fixed.

  Regions in which the person flag is indefinite from time t7 are region A, region B, and region C. First, the processing device 30 executes reverse determination processing for the region A. Here, the person flag in the area A at the reference time t7 is 0, and the motion information is 1. Therefore, the person flag in the area A at time t6 is determined as 1. Here, in the region A, since the person flag at time t5, which is the time immediately before the confirmed time t6, is also uncertain, the reverse determination process is executed with reference to the time t6. The person flag in the area A at the reference time t6 is 1, and the motion information is 1. Therefore, the person flag in the region A at time t5 is not fixed and remains indeterminate. Note that the arrow from time t6 to t5 in the region A shown in the figure and the "X" at the tip of the arrow cannot determine the uncertainty of time t5 in the reverse determination process based on time t6. It shows that.

  Next, the processing device 30 executes reverse determination processing in order for the region B and the region C. The person flag in the region B and the region C at the reference time t7 is 0, and the motion information is 0. Accordingly, the person flags in the area B and the area C at time t6 are both determined to be 0. Here, in the region C, since the person flag at time t5, which is the time immediately before the confirmed time t6, is also uncertain, the reverse determination process is executed with reference to the time t6. The person flag in the region C at the reference time t6 is 0, and the motion information is 1. Accordingly, the person flag in region C at time t5 is determined to be 1.

  In this state, the time when the person flag is uncertain is only the time t5. At time t5, the region where the person flag is indefinite is only the region A. As described above, when there is only one area in which the person flag is indefinite within the same time, the person flag in that area can be determined by the same determination process. Here, when the same determination process is executed at time t5, the person flag in the area A that has been indefinite is fixed to zero.

  In the above-described example, the person flag can be determined in all areas at all times by executing the reverse determination process and the same determination process after executing the sequential determination process and the same determination process. did it. However, even in these processes, when there is an area where the person flag cannot be determined, the number of cases where the flag can be determined can be increased by sequentially executing the sequential determination process, the reverse determination process, and the same determination process again. . Even in such a case, if it is not fixed, the probability that it can be determined can be further increased by shortening the detection cycle in the detection device 10 or the number-of-people acquisition device 20.

  FIG. 14D is a diagram illustrating a human trajectory estimated from the arrangement of each region in FIG. FIG. 14D shows that the persons X and Y have entered the respective rooms and moved to the respective areas at the timing indicated by parentheses. The trajectory of the person X is indicated by a solid line arrow, and the trajectory of the person Y is indicated by a dashed arrow. As shown in the figure, the person X enters the area A at the timing of the time t1, and thereafter moves from the area A → the area B → the area C → the area C → the area C → the area F → the area I from the time t2 to the time t7. Indicates that In addition, the person Y enters the area A at the timing of time t4, and thereafter moves from area A to area B to area A to area D from time t5 to time t7.

As described in the estimation method (4) described above, the following rules can be derived from the trajectory of the person shown in FIG.
○ When all persons on the premises where the monitoring system 100 according to the present embodiment is installed move to an area where the person flag is determined, for example, move to the area D or the area I at the time t7 in the above example. When this is done, the person flag in all areas at the time is determined by the method (2). By confirming the person flag of all the areas at that time, it is possible to escape from indefiniteness by going back to the past by the reverse determination process.

  The present invention has been described based on the embodiments. This embodiment is an exemplification, and various modifications can be made to the combination of the embodiment and modifications, or the combination of each component and each processing process, and such modifications are also applicable to the present invention. It will be understood by those skilled in the art that it is in the range.

  In the embodiment of the present invention, the output unit 16 has been described as outputting information indicating an operation to the processing device 30 at a predetermined cycle. However, the present invention is not limited to this. For example, the processing device 30 notifies the output unit 16 of an output request signal for requesting output of information indicating the operation, and the output unit 16 responds to the output request signal. Information indicating the operation may be output to 30. Further, the output unit 16 stocks information indicating an operation until an output request signal is notified from the processing device 30, and collects the stocked information to the processing device 30 in response to an output request. It may be output. In this case, since the processing device 30 can control the output of the output unit 16, the processing device 30 can be designed regardless of the specifications of the output unit 16, and the design flexibility can be improved.

  Further, as shown in FIG. 11A and FIG. 13A, the shape of the monitored region has been described as a square shape. However, the present invention is not limited to this. For example, the monitored region may be circular or polygonal. Even in these cases, it can be easily thought of by those skilled in the art that the same effect can be obtained.

It is a figure which shows the structural example of the monitoring system concerning embodiment of this invention. It is a figure which shows the structural example of the detection apparatus of FIG. It is a timing chart which shows the input-output example of the infrared detection part of FIG. It is a figure which shows the structural example of the processing apparatus of FIG. It is a state transition diagram which shows the example of the state transition in the monitoring system of FIG. FIG. 6 is a diagram illustrating a state relationship diagram between a current state and a past state in the state transition diagram of FIG. 5. It is a flowchart which shows the operation example of the sequential determination process concerning the processing apparatus of FIG. It is a flowchart which shows the operation example of the same time determination process concerning the processing apparatus of FIG. It is a flowchart which shows the operation example of the reverse time determination process concerning the processing apparatus of FIG. It is a flowchart which shows the operation example of the processing apparatus of FIG. FIG. 11A is a diagram illustrating a region in the first operation example of the embodiment of the present invention. FIG.11 (b) is a figure which shows the example of the motion information concerning each area | region in the 1st operation example of embodiment of this invention. FIG.11 (c) is a figure which shows the example of a transition of the number of persons in the 1st operation example of embodiment of this invention. FIG. 12A is a diagram illustrating an estimation result of the arrangement of each region in the first operation example of the embodiment of the present invention. FIG. 12B is a diagram illustrating a trajectory of a person or the like estimated from the arrangement of each region in FIG. FIG. 13A shows a region in the second operation example of the embodiment of the present invention. FIG. 13B is a diagram illustrating an example of motion information regarding each region in the first operation example of the embodiment of the present invention. FIG. 13C is a diagram illustrating a transition example of the number of persons in the second operation example of the embodiment of the present invention. FIG. 14A is a diagram illustrating a first estimation result of the arrangement of each region in the second operation example of the embodiment of the present invention. FIG. 14B is a diagram illustrating a second estimation result of the arrangement of the areas in the second operation example of the embodiment of the present invention. FIG. 14C is a diagram illustrating a third estimation result of the arrangement of the regions in the second operation example of the embodiment of the present invention. FIG. 14D is a diagram illustrating a human trajectory estimated from the arrangement of each region in FIG.

Explanation of symbols

  S0 state, S1 state, S2 state, S3 state, 10 detection device, 12 infrared detection unit, 16 output unit, 18 initialization unit, 20 number acquisition device, 22 transition, 30 processing device, 32 first reception unit, 34 first 1 memory, 36 second receiving unit, 38 position estimating unit, 40 second memory, 42 control unit, 44 indicating unit, 100 monitoring system, 110 region, 120 entrance / exit region, 200 state transition diagram, 202 first transition, 204 first 2 transition, 206 3rd transition, 208 4th transition, 210 5th transition, 212 6th transition, 214 7th transition, 216 8th transition, 218 9th transition, 220 10th transition, 300 area column, 310 time column , 320 Number of people field, 400 State relationship diagram, 410 Past state column, 420 Current state , 430 transition column.

Claims (18)

A monitoring system comprising a plurality of detection devices that respectively detect movements of an object such as a human or an animal, and a processing device that processes outputs of the plurality of detection devices,
Each of the plurality of detection devices is
A detection unit for detecting presence or absence of movement of the object in the allocated area;
An output unit that outputs detection information by the detection unit to the processing device at a predetermined output cycle;
The processor is
A receiver for receiving detection information output from the plurality of detection devices;
A monitoring system comprising: an estimation unit that estimates a position of an object based on detection information received by the reception unit. The detection device is an infrared sensor that detects a change in infrared rays at a sampling cycle shorter than the output cycle,
When detecting the movement of the object, the detection unit sets a state value of detection information to ON and maintains the ON value, and when the output unit outputs detection information, sets the state value to OFF. The monitoring system according to claim 1, wherein: The processing apparatus further includes a recording unit that records and holds, for each region, detection information received by the receiving unit at a predetermined cycle,
The estimation unit estimates presence / absence of an object in the region using detection information of a certain region recorded in the recording unit and detection information of the same region continuously recorded in the detection information. The monitoring system according to claim 1, wherein:   The estimation unit uses detection information of a certain area recorded in the recording unit at a certain time and detection information of the same area detected before the detection information, and the object at the time of the area is detected. The monitoring system according to claim 3, wherein a sequential determination process for estimating presence / absence of existence is executed.   The estimation unit uses the detection information of a certain region recorded in the recording unit at a certain time and the detection information of the same region detected after the detection information, and the presence of the object at the time of the region The monitoring system according to claim 3, wherein a reverse determination process for estimating presence / absence of a reverse is executed.   The estimation unit uses detection information of a certain area recorded in the recording unit at a certain time and detection information of the same area detected before the detection information, and the object at the time of the area is detected. If the presence / absence of an object cannot be estimated even after executing the forward determination process for estimating the presence / absence of presence, the detection information at the time of the area recorded in the recording unit, and the detection information 4. The monitoring system according to claim 3, wherein a reverse determination process for estimating presence / absence of an object at the time in the area is executed using detection information of the same area detected later.   The estimation unit uses the detection information of a certain region recorded in the recording unit at a certain time and the detection information of the same region detected after the detection information, and the presence of the object at the time of the region If the presence / absence of the target object cannot be estimated even when the reverse determination process for estimating the presence / absence of the object is performed, the detection information at the time of the area recorded in the recording unit, and the detection information 4. The monitoring system according to claim 3, wherein a sequential determination process for estimating presence / absence of an object at the time of the area is executed using detection information of the same area detected before. The processing apparatus further includes a recording unit that records and holds, for each region, detection information received by the receiving unit at a predetermined cycle,
The estimation unit uses the detection information recorded on the area recorded in the recording unit and the detection information recorded on the area around the area to determine whether the object exists in the area or the surrounding area. The monitoring system according to claim 1, wherein the monitoring system is estimated.   The estimation unit uses the detection information of a plurality of areas recorded at the same time in the recording unit to estimate the presence / absence of an object around the arbitrary area at the same time The monitoring system according to claim 8, wherein: The processing apparatus further includes a recording unit that records and holds, for each region, detection information received by the receiving unit at a predetermined cycle,
The estimation unit uses the detection information of a certain region recorded in the recording unit at a certain time and the detection information of the same region detected after the detection information, and the presence of the object at that time in that region Using the detection information at a certain time of a certain region recorded in the recording unit and the detection information of the same region detected before the detection information If the presence / absence of the object cannot be estimated even after executing the sequential determination process for estimating the presence / absence of the object at that time, a plurality of times recorded in the recording unit at the same time as the time 3. The monitoring system according to claim 1, wherein a simultaneous determination process for estimating presence / absence of an object at the time in an area around an arbitrary area is executed using detection information of the area. The processing apparatus further includes a recording unit that records and holds, for each region, detection information received by the receiving unit at a predetermined cycle,
The estimation unit uses the detection information of a plurality of areas recorded at the same time in the recording unit to estimate the presence / absence of an object at the time in an area around an arbitrary area. If the presence / absence of the object cannot be estimated even after executing the above, the detection information at a certain time recorded in the recording unit and the detection information of the same area detected after the detection information are obtained. Using the reverse determination process for estimating the presence or absence of an object at the time in the area, or the detection information at the time of the area recorded in the recording unit, and recording the detection information before the detection information. 3. The monitoring system according to claim 1, wherein a sequential determination process for estimating presence / absence of an object at the time of the region is executed using the detected information of the same region. The processing apparatus further includes a recording unit that records and holds, for each region, detection information received by the receiving unit at a predetermined cycle,
The estimation unit detects detection information at a first time of a certain region recorded in the recording unit and before the detection information until the presence / absence of an object in all regions at all times is estimated. Using the detection information of the detected same area, a sequential determination process for estimating the presence / absence of an object at the first time in the area, and a plurality of areas recorded at the second time in the recording unit Using the detection information, the same time determination process for estimating the presence or absence of an object in an area around an arbitrary area, the detection information at a third time of a certain area recorded in the recording unit, and the detection information A plurality of determination processes are sequentially performed among the three determination processes including the reverse determination process for estimating the presence / absence of an object at the third time using the detection information of the same area detected later. Characterized by performing Monitoring system according to claim 1 or 2.   The monitoring system according to claim 12, wherein the estimation unit is repeatedly executed in the order of sequential determination processing, same-time determination processing, reverse-time determination processing, and same-time determination processing.   The monitoring system according to any one of claims 1 to 13, wherein when the estimation unit cannot estimate the presence or absence of an object, the output unit sets the output cycle to be short.   The monitoring system according to claim 14, wherein the output unit sets the output cycle to be shorter as the number of objects in the plurality of regions increases. The processing device notifies the output unit of an output request signal for requesting output,
The monitoring system according to claim 1, wherein the output unit outputs detection information to the processing device in response to the notified output request signal.   The output unit holds detection information from the detection unit until an output request signal is notified. When the output request signal is notified from the processing device, the output unit outputs the held detection information to the processing device. The monitoring system according to claim 16.   Detecting movements of objects such as humans or animals in a plurality of regions, storing each detected detection information in a predetermined cycle, and based on each stored detection information, A monitoring method characterized by estimating a position of an object.

Images