JP7334512B2 - Verification method, verification device, and program - Google Patents

Description

The present disclosure relates to a method, apparatus, and program for collating feature amounts of moving objects.

Techniques for tracking a moving object using images obtained from imaging devices have been developed. For example, there is a technique for extracting feature amounts of a moving object from videos, matching the feature amounts between videos, identifying the moving object, and tracking the moving object.

Here, when a moving object moves from the imaging range of one imaging device to the imaging range of another imaging device, even if the same moving object is the same moving object, it is different from the moving object existing in the imaging range of the one imaging device. , there is a possibility that a moving object existing in the imaging range of another imaging device may be recognized as a different moving object.

On the other hand, Patent Literature 1 discloses a technique for identifying a person appearing in one imaging range and a person appearing in another imaging range. The human tracking device of Patent Document 1 records all the feature amounts of persons appearing in a plurality of imaging ranges, and when a person appears in one imaging range, the person's feature amount and , is compared with the recorded features. As a result of collation, if there is a matching feature amount, the person tracking device of Patent Literature 1 determines whether the collation result is correct.

JP 2006-236255 A

In order to identify a moving object, the human tracking device of Patent Document 1 compares the feature amount of a person appearing in one imaging range with all recorded feature amounts. Therefore, as the number of recorded features increases, there is a risk that computational resources will increase.

The present disclosure has been made in view of the above problems, and a main object of the present disclosure is to provide a matching method or the like that suppresses an increase in computational resources in identifying a moving object that has appeared in an imaging range.

A collation method according to an aspect of the present disclosure includes each of a plurality of predetermined regions within an imaging range and a corresponding region that is one of the plurality of regions and is associated with each region. selecting the corresponding area associated with the area in which the first moving object existed from among the plurality of areas from the area correspondence information included therein, and selecting the area in which the second moving object existed from among the plurality of areas and a feature quantity of the second moving body that is extracted when the second moving body exists in the region and that is associated with the region, the moving body information containing: Acquiring a feature amount associated with the selected corresponding region, and comparing the acquired feature amount with the feature amount of the first moving object extracted when the first moving object exists. do.

A matching device according to an aspect of the present disclosure includes each of a plurality of regions predetermined within an imaging range and a corresponding region that is one of the plurality of regions and is associated with each of the regions. area selection means for selecting, from among the plurality of areas, the corresponding area associated with the area in which the first moving object existed, from the area correspondence information included; and a feature quantity extracted when the second moving body exists in the region, the feature quantity of the second moving body associated with the region. A feature amount associated with the selected corresponding region is acquired from the information, and the acquired feature amount and the feature of the first moving object extracted when the first moving object exists. and a collating means for collating the quantity.

A program according to an aspect of the present disclosure includes each of a plurality of regions predetermined within an imaging range and a corresponding region that is one of the plurality of regions and is associated with each of the regions. a process of selecting, from the area correspondence information, the corresponding area associated with the area in which the first moving object existed among the plurality of areas; From moving object information including an area and a feature amount of the first moving object that is extracted when the second moving object is present in the area and that is associated with the area, acquiring a feature amount associated with the selected corresponding region, and combining the acquired feature amount with the feature amount of the second moving object extracted when the second moving object exists; A computer is made to perform a process to collate.

Advantageous Effects of Invention According to the present disclosure, it is possible to suppress an increase in computational resources in identifying a moving object that has appeared in an imaging range.

It is a block diagram which shows an example of the hardware constitutions of the computer apparatus which implement|achieves the collation apparatus in each embodiment. It is a figure showing typically an example of composition of a tracking system concerning a 1st embodiment. It is a block diagram showing an example of functional composition of a tracking system concerning a 1st embodiment. It is a figure showing an example of functional composition of a collation device concerning a 1st embodiment. 4 is a diagram showing an example of area correspondence information according to the first embodiment; FIG. It is a figure which shows an example of mobile body information concerning 1st Embodiment. 4 is a flowchart for explaining the operation of the imaging device according to the first embodiment; 4 is a flowchart for explaining the operation of the collation device according to the first embodiment; FIG. 5 is a diagram showing an example of updated mobile object information according to the first embodiment; FIG. 5 is a flowchart for explaining matching processing of the matching device according to the first embodiment; It is a block diagram showing an example of functional composition of a tracking system concerning modification 1 of a 1st embodiment. It is a block diagram showing an example of functional composition of a tracking system concerning modification 2 of a 1st embodiment. It is a block diagram showing an example of functional composition of a tracking system concerning modification 3 of a 1st embodiment. FIG. 11 is a diagram showing an example of mobile object information according to the second embodiment; FIG. It is a block diagram showing an example of functional composition of a tracking system concerning a 3rd embodiment. It is a figure which shows an example of the operating information concerning 3rd Embodiment.

Hardware configuration example of each embodiment Hardware configuring a collation device according to each embodiment will be described. FIG. 1 is a block diagram showing an example of the hardware configuration of a computer that implements a matching device in each embodiment. Each block shown in FIG. 1 can be realized by a combination of software and a computer device 10 that realizes the matching device and matching method in each embodiment.

As shown in FIG. 1 , computer device 10 includes processor 11 , RAM (Random Access Memory) 12 , ROM (Read Only Memory) 13 , storage device 14 , input/output interface 15 , bus 16 , and drive device 17 .

Storage device 14 stores program 18 . The processor 11 uses the RAM 12 to execute a program 18 related to this verification device. Specifically, for example, the program 18 includes a program that causes a computer to execute the processes shown in FIGS. 7 and 8. FIG. As the processor 11 executes the program 18, the function of each component of the collation device (region selection unit 310 and collation unit 320, which will be described later) is realized. Program 18 may be stored in ROM 13 . Further, the program 18 may be recorded on the recording medium 20 and read using the drive device 17, or may be transmitted from an external device (not shown) via a network (not shown).

The input/output interface 15 exchanges data with peripheral devices (keyboard, mouse, display device, etc.) 19 . The input/output interface 15 functions as means for acquiring or outputting data. A bus 16 connects each component.

It should be noted that there are various modifications of the implementation method of the matching device. For example, the verification device can be implemented as a dedicated device. Also, the matching device can be realized based on a combination of a plurality of devices.

A processing method of recording a program for realizing each component in the function of each embodiment on a recording medium, reading the program recorded on the recording medium as a code, and executing it on a computer is also included in the scope of each embodiment. . That is, a computer-readable recording medium is also included in the scope of each embodiment. In addition to the recording medium on which the above program is recorded, the program itself is also included in each embodiment.

The recording medium is, for example, a floppy (registered trademark) disk, hard disk, optical disk, magneto-optical disk, CD (Compact Disc)-ROM, magnetic tape, non-volatile memory card, or ROM, but is not limited to these examples. In addition, the program recorded on the recording medium is not limited to a program that executes processing by itself, but in cooperation with other software and functions of an expansion board, it runs on an OS (Operating System) to execute processing. A program for executing the program is also included in the category of each embodiment.

First Embodiment Next, each component of a tracking system including a matching device will be described.

FIG. 2 is a diagram schematically showing an example of the configuration of the tracking system 1000 according to the first embodiment. As shown in FIG. 2 , tracking system 1000 includes imaging device 100 , tracking device 200 , matching device 300 , and storage device 400 . The tracking system 1000 is a system that tracks a moving object included in a plurality of images captured by the imaging device 100 based on at least the tracking device 200 and the verification device 300 .

Figure 2 also shows the tracking system 1000 installed in the following situation. The moving object shown in FIG. 2 uses an elevator to move from the first floor to the second floor of a building. Here, the 1st, 2nd, and 3rd floors respectively indicate the floors of the building, and the 1st floor a, the 1st floor b, the 2nd floor a, and the 3rd floor a indicate the rooms on each floor. 1st floor a, 2nd floor a, and 3rd floor a are connected via an elevator. 1st floor a and 1st floor b are connected through a doorway. The arrows shown in FIG. 2 indicate the movement of the moving object. At this time, the moving body moves from point A on the first floor to point B, and uses an elevator to move to point C on the second floor.

The imaging device 100 is installed on each floor from the first floor to the third floor, and images each room. Henceforth, the imaged image|video is also called imaging data. The tracking device 200 generates flow line information indicating the motion or movement trajectory of the moving body included in the imaging data from the imaging data, and tracks the moving body. The collation device 300 collates the feature amount of a moving object included in one image with the feature amount of a moving object included in another image.

In the situation shown in FIG. 2, the tracking device 200 tracks the moving object while the moving object moves from point A to point B on the first floor. A point A is a point on the first floor a, and a point B is a point on the first floor a near the elevator. Each of the points A and B exists within a predetermined area within the imaging range. In each embodiment including this embodiment, an example of a predetermined area within the imaging range is a boundary area. The boundary area is an area within the imaging range of the imaging device 100 and near the boundary between the inside of the imaging range and the outside of the imaging range (in the situation of FIG. 2, the outside of the imaging range is inside the elevator). When the moving object moves to the boundary region x2 including the point B, the imaging device 100 extracts the feature amount of the moving object being tracked from the video. Then, the matching device 300 associates the extracted feature quantity with the information indicating the boundary region x2 and stores them as moving object information (details of which will be described later).

Further, when the moving object moves from inside the elevator, which is outside the range captured by the imaging device 100, to the boundary region x3 including the point C, the matching device 300 corresponds to each of a plurality of pre-stored boundary regions. The boundary area corresponding to the boundary area x3 is selected from the area correspondence information (details will be described later) including the area (boundary area) to be crossed as information. Here, when the moving body moves from the inside of the elevator to the boundary area x3, the movement path of the moving body is either from the boundary area x2 or the boundary area x4 using the elevator, or from the boundary area x3 to the inside of the elevator. It is limited to moving and moving again to the boundary area x3. In the area correspondence information, such a certain boundary area is defined as a boundary area in which the moving object can move via outside the imaging range. In the present embodiment, a boundary area in which a moving object can move through a boundary area outside the imaging range is referred to as a "corresponding boundary area" or a "corresponding area." In the situation of FIG. 2, matching device 300 selects boundary region x2, boundary region x3 and boundary region x4.

Then, the collation device 300 collates the feature amount of the moving object that has moved to the boundary area x3 with the feature amount included in the moving object information and associated with the selected boundary area. In the situation of FIG. 2, the feature quantity to be matched with the feature quantity of the moving object that has moved to the boundary region x3 is the feature quantity associated with the selected boundary region x2, boundary region x3, or boundary region x4. As a result, the verification device 300 identifies the moving object that has moved to the boundary area x3 and the moving object that has existed in the boundary area x2. The tracking device 200 tracks the moving object that has moved to the point C as the same moving object that has moved from the point A to the point B. FIG.

In addition, in this embodiment, the moving object is, for example, a person, but is not limited to this. The moving object may be an automobile or the like.

FIG. 3 is a block diagram showing an example of the functional configuration of the tracking system 1000 shown in FIG. 2. As shown in FIG. As shown in FIG. 3 , the imaging device 100 includes an imaging unit 110 , a moving body detection unit 120 , an area determination unit 130 and a feature amount extraction unit 140 . The imaging device 100 captures an image of a predetermined range, and transmits information on the moving object detected from the captured data to the tracking device 200 , the verification device 300 and the storage device 400 .

The tracking device 200 uses the information received from the imaging device 100 and the verification device 300 to track the moving object detected from the imaging data. Tracking device 200 also transmits information indicating whether or not the detected moving object is being tracked to verification device 300 .

FIG. 4 is a block diagram showing an example of the functional configuration of the matching device 300. As shown in FIG. As shown in FIG. 4 , collation device 300 includes area selection section 310 and collation section 320 . The matching device 300 uses the information received from the imaging device 100 and the tracking device 200 and the information stored in the storage device 400 to determine the feature amount of the moving object included in one imaging data and the other imaging data. is collated with the feature quantity of the moving object included in .

Returning to FIG. 3 , the storage device 400 includes an area correspondence information storage section 410 and a moving object information storage section 420 . The area correspondence information storage unit 410 stores area correspondence information, and the mobile unit information storage unit 420 stores mobile unit information.

Each block of the imaging device 100, the matching device 300, and the storage device 400 shown in FIGS. 3 and 4 may be implemented in a single device, or may be implemented separately in a plurality of devices. Also, data transfer between blocks may be performed via any means such as a data bus, a network, or a portable storage medium.

Details of Imaging Apparatus 100 Next, each component of the imaging apparatus 100 will be described. In FIG. 3, the imaging unit 110 images a predetermined range. A predetermined range captured by the imaging unit 110 is referred to as an “imaging range”. In the situation shown in FIG. 2, the imaging range is each room except the inside of the elevator and between the entrances and exits of the rooms in which the imaging devices 100 are respectively installed. The imaging unit 110 continues imaging the imaging range until a predetermined end instruction is given. Here, the video imaged by the imaging unit 110 may be each of a plurality of frames forming a moving image, or may be a still image. In this specification, the image captured by the imaging device 100 is referred to as "captured data". The imaging data includes color information or pixel value data of each pixel of a frame or a still image, as well as the time when the imaging data was generated, information on the location where the imaging device 100 that captured the imaging data is installed, and the like. may be included.

The moving object detection unit 120 sequentially reads the imaging data generated by the imaging unit 110 and detects a moving object included in the read imaging data. For the detection of the moving object, for example, a method of detecting the moving object based on the difference between a plurality of pieces of image data, that is, the background subtraction method can be used, but the method is not limited to this example. Various known methods such as a method of detecting a human figure may be used as a method of detecting a moving object. Further, the moving body detection unit 120 detects position information in the imaging range for the detected moving body.

The area determination unit 130 determines whether or not the moving object exists in the boundary area based on the position information detected by the moving object detection unit 120 . In the situation shown in FIG. 2, there are multiple boundary regions. For example, the hatched area near the elevators on each floor and the hatched area near the entrances to the first floor a and the first floor b are boundary areas. Note that the boundary area is not limited to this example. For example, the boundary area may be near the entrance of an escalator, near the entrance of a vehicle such as a train, or near the entrance of a restroom, fitting room, or the like. Alternatively, the boundary area may be near the entrances and exits of tunnels, underpasses, and the like. In this embodiment, each boundary region is distinguished. If the moving object exists in the boundary area, the area determination unit 130 identifies in which boundary area the moving object is located. Hereinafter, information including information indicating whether or not a moving object exists in the boundary area and, if the moving object exists in the boundary area, information identifying the boundary area in which the moving object exists is referred to as "area "identification information".

When a moving object is detected from the imaging data, the feature quantity extraction unit 140 extracts the feature quantity of the detected moving object. The feature amount to be extracted may be, for example, a feature amount related to the color, outline, etc. of the moving object. When the moving object is a person, the feature amount may be information indicating features related to the person's face or iris.

The imaging device 100 transmits the position information of the moving object, the area identification information about the moving object, and the feature amount of the moving object to the tracking device 200 and the matching device 300 . In this embodiment, the moving object detection unit 120, the area determination unit 130, and the feature amount extraction unit 140 transmit the position information, the area identification information, and the feature amount of the moving object, but the present invention is not limited to this example. For example, a communication unit capable of communicating with the tracking device 200 and the matching device 300 may be provided in the imaging device 100, and the communication unit may transmit each piece of information.

Further, the information that the imaging device 100 transmits to the tracking device 200 and the matching device 300 is not limited to the above information. The imaging device 100 may transmit the imaging data or the time when the imaging data was generated in addition to the above information. Further, when a plurality of moving objects are detected in one imaging data, the imaging device 100 may transmit feature amounts, position information, and area identification information regarding the plurality of moving objects.

Details of Tracking Device 200 Next, the tracking device 200 will be described.

The tracking device 200 receives the feature quantity, position information, and area identification information of the detected moving object from the imaging device 100 . The tracking device 200 tracks the moving object included in the continuous imaging data according to the received information. For example, the tracking device 200 compares the feature values continuously received from the same imaging device 100, and if there is a matching feature value, the position information is used to track the movement of the mobile object. to generate flow line information indicating the trajectory of If a feature amount that matches the feature amount of the moving object being tracked is not received from the same imaging device 100, the moving object being tracked moves from one imaging range to outside the one imaging range. show. An identification ID (IDentification) generated by the verification device 300 is associated with the moving object being tracked. An identification ID is information for identifying a mobile object. In the present embodiment, a moving object that has moved from outside the imaging range to the boundary area, that is, a moving object that the tracking device 200 has not started tracking is referred to as a “non-tracking moving object” or a “first moving object”. Further, in the present embodiment, the moving object that the tracking device 200 continues to track or has been tracking until it moves out of one imaging range is referred to as a "tracking moving object" or a "second moving object." ”.

When the moving object exists in the boundary area, that is, when the area identification information includes information indicating that the moving object exists in the boundary area, the tracking device 200 detects that the moving object is It is determined whether the mobile object is not being tracked or is being tracked. The tracking device 200 transmits a notification indicating the determination result to the verification device 300 . For example, if the moving object existing in the boundary area is the moving object being tracked, a notification including the identification ID of the moving object and the determination result is transmitted to the verification device 300 . If the moving object existing in the boundary area is not being tracked, tracking device 200 transmits a notification including the determination result to matching device 300 .

If the moving object existing in the boundary area is not being tracked, the tracking device 200 associates the identification ID included in the notification received from the verification device 300 and tracks the moving object.

In this embodiment, the tracking device 200 may use the feature quantity extracted by the feature quantity extraction unit 140 of the imaging device 100 when tracking the moving object, or may receive imaging data from the imaging device 100, The tracking device 200 may extract the feature quantity, and the moving object may be tracked using the feature quantity extracted by the tracking device 200 . Also, the method of tracking a moving object is not limited to this, and various known methods may be used.

Details of Verification Device 300 Next, each component of the verification device 300 will be described.

In FIG. 4, the area selection unit 310 is an example of area selection means. The area selection unit 310 selects one of each of a plurality of areas (in this embodiment, a plurality of boundary areas) predetermined within the imaging range and a plurality of areas, and is associated with each area. The corresponding area associated with the area in which the first moving object was present is selected from the area correspondence information including the corresponding area. For example, it is assumed that the imaging device 100 detects a moving object that has moved from inside the elevator to the boundary region x3 in the situation of FIG. The region selection unit 310 receives the region identification information about the boundary region x3 from the imaging device 100, and from the tracking device 200, the moving body detected in the boundary region x3 is a moving body not being tracked (first moving body). Receive a notification that there is Furthermore, area selection section 310 selects the corresponding area corresponding to boundary area x3 indicated by the area identification information from the area correspondence information stored in area correspondence information storage section 410 of storage device 400 . Specifically, area selection section 310 reads information indicating a corresponding area corresponding to boundary area x3 from area correspondence information storage section 410 of storage device 400 .

FIG. 5 is a diagram showing an example of area correspondence information. As shown in FIG. 5, the area correspondence information is information in which a corresponding boundary area is defined for each boundary area among a plurality of boundary areas.

The matching unit 320 compares a region in which the second moving object existed among a plurality of regions (a plurality of boundary regions in this embodiment) and a feature amount extracted when the second moving object exists in the region. Then, the feature quantity associated with the selected corresponding region is acquired from the moving body information including the feature quantity of the second moving body associated with the region. The second moving object is the moving object that the tracking device 200 continues to track or has been tracking. Specifically, matching section 320 receives information indicating the corresponding area selected by area selecting section 310 . Furthermore, the matching unit 320 extracts the feature amount of the moving object associated with the selected corresponding region from the moving object information stored in the moving object information storage unit 420 of the storage device 400 .

FIG. 6 is a diagram showing an example of mobile information. As shown in FIG. 6, the moving body information is associated with an identification ID, a feature amount of the moving body, and information for identifying a boundary area. The feature amount and the information identifying the boundary area included in the moving object information are the feature amount extracted when the moving object was present in the boundary area and the information identifying the boundary area where the moving object was present. show. For example, the second line of the moving body information shown in FIG. 6 indicates that the feature quantity extracted when the moving body with the identification ID “001” exists in the “boundary region x1” is the “feature quantity a”. ing.

An example will be described in which the collating unit 320 acquires the feature amount associated with the corresponding area selected by the area selecting unit 310 from the mobile object information. When the selected corresponding areas are the boundary area x1 and the boundary area x5, the matching unit 320 determines the feature amounts a, and feature quantity b. On the other hand, in the situation of FIG. 2, when a moving object (first moving object) that is not being tracked exists in the boundary region x3, the region selected by the region selection unit 310 as the corresponding region corresponding to the boundary region x3 is the boundary A region x2, a boundary region x3, and a boundary region x4. Suppose that in the moving object information, the feature amount of a second moving object different from the first moving object is stored for the boundary regions x3 and x4, and the characteristic amount of the second moving object that is the same as the first moving object is stored for the boundary region x2. Assume that feature values are stored. In this case, the collation unit 320 compares the feature amount of the second moving object different from the first moving object associated with the boundary areas x3 and x4, and the feature amount of the first moving object associated with the boundary area x2. A feature amount of a second moving body that is the same as the body is acquired.

And the collation part 320 collates the acquired feature-value with the feature-value of the 1st moving body extracted when the 1st moving body existed. As a result of matching, there is a matching feature amount means that a moving object (first moving object) included in one imaging data and a moving object (second moving object) included in other imaging data are Indicates that it has been identified. As a result of matching, if there is a matching feature quantity, the matching unit 320 transmits to the tracking device 200 a matching result including the identification ID of the moving object associated with the matching feature quantity. At this time, the matching unit 320 associates the identification ID, the feature amount of the first moving object, and the information of the boundary area where the first moving object exists, and updates the moving object information. Collation unit 320 is an example of collation means.

As a result of matching, if there is a matching feature quantity, the matching unit 320 may discard the feature quantity that matches the feature quantity of the first moving object from the moving object information. In the situation of FIG. 2, when the feature quantity of the moving body that has moved to the boundary region x3 matches the feature quantity associated with the boundary region x2 included in the moving body information, the collating unit 320 determines from the moving body information: , the feature quantity associated with the boundary region x2 may be discarded.

In addition, the matching unit 320 determines that the feature amount of the first moving body, which is associated with the boundary area in which the first moving body was present, is associated with the boundary area that is not associated with the moving body information. , the feature amount may be discarded from the moving body information. For example, the collating unit 320 discards from the moving object information the feature amount in which the identification ID that is the same as the identification ID of the first moving object and the boundary area that is different from the corresponding area selected by the area selection unit 310 are associated with each other. You may In the situation of FIG. 2, when the moving object information contains the same moving object feature quantity as the moving object that has moved to the boundary region x3 and is associated with the boundary region x1 or the boundary region x5, The matching unit 320 may discard the feature amount of the moving object associated with the boundary area x1 or the boundary area x5 from the moving object information. Thus, by discarding the feature amount from the mobile object information, it is possible to suppress an increase in the amount of data of the mobile object information. Note that the moving object information storage unit 420 may have the function of discarding the feature amount.

As a result of matching, if there is no matching feature amount, the detected moving object is the moving object that has moved into the imaging range for the first time. Therefore, the collation unit 320 generates a new identification ID for the moving object from which the imaging device 100 has been detected. The collation unit 320 then transmits the collation result including the generated identification ID to the tracking device 200 . At this time, the collation unit 320 associates the generated identification ID, the feature amount of the detected moving object, and the information of the boundary area where the detected moving object exists, and updates the moving object information.

In addition, when receiving from the tracking device 200 a notification that the detected moving object is the moving object being tracked, the matching unit 320 matches the identification ID of the moving object included in the notification with the feature amount acquired from the imaging device 100 . and information indicating the boundary area where the detected moving object existed, and update the moving object information.

In this embodiment, the notification received by the verification device 300 from the tracking device 200 is received by either the area selection unit 310 or the verification unit 320 from the tracking device 200, and even if one of them sends the notification to the other. Alternatively, both may receive from tracking device 200 . Also, the verification device 300 is provided with a communication unit capable of communicating with the tracking device 200, the communication unit receives the notification, and the communication unit notifies either the area selection unit 310 or the verification unit 320 according to the content of the notification. may be sent.

Operation of Tracking System 1000 Next, the operation of tracking system 1000 will be described. In the tracking system 1000 of this embodiment, the imaging device 100 detects a moving object from imaging data and transmits information about the detected moving object to the tracking device 200 and the matching device 300 . Tracking device 200 tracks the moving object based on the received information. When the detected moving object is the moving object being tracked (second moving object) and exists in the boundary area, the verification device 300 compares the feature amount of the detected moving object and the presence of the moving object. and the information of the boundary area that has been obtained are stored in association with each other. When the detected moving object is a moving object that is not being tracked (first moving object) and is present in the boundary area, the verification device 300 determines the feature quantity of the detected moving object and the The moving object is identified by collating with the feature values provided. At this time, the matching device 300 narrows down objects to be matched with the feature amount of the detected moving object from the moving object information using the information of the boundary area where the detected moving object exists.

Operations of the imaging device 100 and the matching device 300 will be described below with reference to the flowcharts of FIGS. 7 and 8. FIG. In the operation described below, it is assumed that the area correspondence information shown in FIG. 5 is stored in the area correspondence information storage unit 410, and the mobile information shown in FIG. do. In this embodiment, the operations of the imaging device 100, the tracking device 200, and the matching device 300 will be described using the situation shown in FIG. 2, but the scene to which the tracking system 1000 is applied is not limited to this example. Also, in this specification, each step of the flowchart is expressed using a number attached to each step, such as "S701".

First, the operation of the imaging device 100 will be described.

FIG. 7 is a flowchart for explaining the operation of the imaging device 100. FIG. First, the imaging unit 110 captures an image of an imaging range and generates imaging data (S701). Here, as shown in FIG. 2, the imaging device 100 is installed in each room on each floor, and each room except the inside of the elevator and between the entrances and exits is the imaging range.

The moving object detection unit 120 reads the generated imaging data. When the moving object detection unit 120 detects a moving object from the imaging data (“YES” in S702), the moving object detection unit 120 sends a notification indicating that the moving object has been detected to the area determination unit 130 and the feature amount extraction unit 140 . At this time, the notification includes the location information of the moving object acquired by the moving object detection unit 120 . If the moving object detection unit 120 does not detect a moving object from the imaging data ("NO" in S702), the imaging device 100 returns to the process of S701.

Upon receiving the notification from the moving body detection unit 120, the feature amount extraction unit 140 extracts the feature amount of the detected moving body (S703). Upon receiving the notification from the moving object detection unit 120, the area determination unit 130 determines whether or not the detected moving object exists within the boundary area based on the position information included in the notification. If the area determining unit 130 determines that the moving object is within the boundary area (“YES” in S704), it specifies in which boundary area the moving object is located (S705). For example, when the position information included in the notification from the moving object detection unit 120 indicates point B, the region determination unit 130 determines that the moving object exists within the boundary region x2. Then, area determination section 130 generates area identification information.

The moving object detection unit 120, the feature amount extraction unit 140, and the area determination unit 130 transmit the position information, the feature amount, and the area identification information acquired in S702, S703, and S704 to the tracking device 200 and the matching device 300. (S706).

When the area determination unit 130 determines that the moving object is outside the boundary area (“NO” in S704), the area identification information includes information indicating that the moving object is outside the boundary area. . Then, the moving object detection unit 120, the feature amount extraction unit 140, and the region determination unit 130 transmit the acquired information to the tracking device 200 (S707).

Next, the operation of the matching device 300 when information is received from the imaging device 100 will be described.

FIG. 8 is a flow chart for explaining the operation of the collation device 300. As shown in FIG. First, the matching device 300 receives the feature amount of the moving object detected from the imaging device 100 and the region identification information (S801). Next, verification device 300 receives a notification including information indicating whether or not the detected moving object is being tracked from tracking device 200 (S802). If the detected moving object is the moving object being tracked ("YES" in S803), that is, if the detected moving object is the second moving object, the verification unit 320 checks the identification ID included in the notification. Then, the moving body information is updated based on the feature amount of the detected moving body and the area identification information (S804). For example, it is assumed that the information of “characteristic quantity e” and “boundary region x2” is received from the imaging device 100 and the identification ID “005” is received from the tracking device 200 . The verification unit 320 then accesses the mobile unit information storage unit 420 of the storage device 400 and updates the mobile unit information using the received information. FIG. 9 is a diagram showing an example of mobile information obtained by updating the mobile information shown in FIG. As shown in FIG. 9, the collation unit 320 associates the identification ID "005", the "feature value e", and the "boundary region x2", and updates the mobile information by adding the mobile information.

If the detected moving object is not being tracked (“NO” in S803), for example, if it is the first moving object that has moved from the elevator to the boundary area x3 as shown in FIG. performs collation processing (S805).

FIG. 10 is a flowchart for explaining the matching process of the matching device 300. As shown in FIG. Area selection unit 310 reads area correspondence information from area correspondence information storage unit 410 of storage device 400 . Then, the area selection unit 310 selects a corresponding area corresponding to the boundary area indicated by the area identification information from the area correspondence information (S1001). Area selection section 310 sends information on the selected corresponding area to matching section 320 . For example, when the area identification information includes information indicating "boundary area x3", area selection section 310 selects "boundary area x2”, “boundary region x3”, and “boundary region x4”. Then, information indicating “boundary region x2”, “boundary region x3”, and “boundary region x4” is sent to collating section 320 .

The matching unit 320 receives the feature amount from the imaging device 100 and the corresponding area information from the area selection unit 310 , and then reads the moving object information from the moving object information storage unit 420 of the storage device 400 . Then, the matching unit 320 acquires the feature amount associated with the selected corresponding area from the mobile object information (S1002). Then, the collation unit 320 collates the acquired feature amount with the feature amount of the first moving object (S1003).

Returning to FIG. 8, if there is a matching feature amount as a result of matching (“YES” in S806), the matching unit 320 extracts the matching result including the identification ID associated with the matching feature value from the moving object information. It is transmitted to the tracking device 200 (S807). Then, the matching unit 320 updates the moving object information based on the identification ID associated with the matching feature amount, the detected moving object feature amount, and the region identification information (S808). For example, assume that the feature quantity received from the imaging device 100 is “feature quantity e”. When the information of the corresponding areas received from the area selection unit 310 is information indicating the boundary area x2, the boundary area x3, and the boundary area x4, the moving object information shown in FIG. Quantities "feature e", "feature d", and "feature c" are obtained. Then, it is found that the feature amount that matches the feature amount received from the imaging device 100 is the feature amount associated with the identification ID "005". Therefore, the collation unit 320 transmits the collation result including the identification ID “005” to the tracking device 200 . Then, the matching unit 320 associates the identification ID “005”, the “feature value e”, and the “boundary region x3” included in the region identification information with each other and adds them to the moving body information, thereby obtaining the moving body information. to update. The tracking device 200 starts tracking the detected moving body as the moving body with the identification ID "005".

As a result of matching, if there is no matching feature amount ("NO" in S806), the matching unit 320 generates a new identification ID, and transmits the matching result including the generated identification ID to the tracking device 200 (S809). ). The matching unit 320 updates the mobile object information based on the generated identification ID, the detected feature amount of the mobile object, and the region identification information. The tracking device 200 assigns a newly generated identification ID to the detected moving body and tracks it.

As described above, the matching device 300 according to the first embodiment can be either each of the plurality of regions predetermined within the imaging range or the plurality of regions, and the The corresponding area associated with the area in which the first moving object appeared is selected from among the plurality of areas from the area correspondence information including the corresponding area. Then, the matching device 300 determines, among the plurality of areas, an area in which the second moving object exists, and a feature amount extracted when the second moving object exists in the area, which corresponds to the area. a feature amount associated with the selected corresponding region is acquired from the moving body information including the feature amount of the second moving body attached thereto, and the acquired feature amount and the first moving body appear Then, the feature amount of the first moving object extracted at the time is collated. With this configuration, the collation device 300 can narrow down the objects to be collated with the detected feature amount of the moving object. Therefore, the collation device 300 according to the first embodiment has the effect of making it possible to suppress an increase in computational resources in identifying a moving object appearing in the imaging range.

In addition, since the matching device 300 according to the first embodiment can narrow down the objects to be matched with the detected feature quantity of the moving object, it is possible to improve the accuracy of matching.

[Modification 1]
The collation device may include an area correspondence information storage unit and a moving body information storage unit.

FIG. 11 is a block diagram showing an example of the configuration of a tracking system 1001 according to this modification. As shown in FIG. 11, the configuration of a tracking system 1001 according to this modified example is the tracking system shown in FIG. Similar to system 1000 .

The operation of the tracking system 1001 according to this modification is the same as the flowcharts shown in FIGS. 7, 8 and 10. FIG.

The collation device 301 according to this modification has the same effects as the collation device 300 described in the first embodiment.

[Modification 2]
The imaging device may include an area selection section and a matching section.

FIG. 12 is a block diagram showing an example configuration of a tracking system 1002 according to this modification. As shown in FIG. 12, the configuration of a tracking system 1002 according to this modification includes an imaging device 101, a tracking device 200, and a storage device 400. The imaging device 101 includes an area selection unit 310 and a matching unit 320. .

The operation of the tracking system 1002 according to this modification will be described. The description of the operation of the tracking system 1002 shown in FIG. 12 that overlaps with the description of the first embodiment will be omitted.

The imaging apparatus 102 performs the same processing as S701 to S705 and S707 in the flowchart shown in FIG. After the process of S705, the moving object detection unit 120, the feature amount extraction unit 140, and the area determination unit 130 transfer the position information, the feature amount, and the area identification information acquired in S702, S703, and S704 to the tracking device 200. , and to the area selection unit 310 and the matching unit 320 included in the imaging device 102 .

The imaging device 102 then receives a notification from the tracking device 200 that includes information indicating whether the detected moving object is being tracked. The area selection unit 310 and the matching unit 320 included in the imaging device 102 combine information received by the imaging device 102 from the tracking device 200, information received from the moving object detection unit 120, the feature amount extraction unit 140, and the area determination unit 130. are used to perform the processing from S803 onwards.

The imaging device 101 according to this modification has the same effects as the matching device 300 described in the first embodiment.

[Modification 3]
The matching device may include a moving body detection section, an area determination section, and a feature amount extraction section.

FIG. 13 is a block diagram showing an example of the configuration of a tracking system 1003 according to this modification. As shown in FIG. 13, in the configuration of the tracking system 1003 according to this modification, the imaging device 102 does not include the moving object detection unit 120, the region determination unit 130, and the feature amount extraction unit 140, and the matching device 302 The tracking system 1000 is the same as the tracking system 1000 shown in FIG.

The operation of the tracking system 1003 according to this modification will be described. The description of the operation of the tracking system 1003 shown in FIG. 13 that overlaps with the description of the first embodiment will be omitted.

The imaging device 102 generates imaging data and transmits the generated imaging data to the matching device 302 .

When the matching device 302 receives the imaging data, the moving object detection unit 120 performs the processing of S702, the feature amount extraction unit 140 performs the processing of S703, and the area determination unit 130 performs the processing of S704 and S705. Next, the collation device 302 transmits the feature quantity, position information, and area identification information of the detected moving object to the tracking device 200 .

Then, if the detected moving body exists in the boundary area, the verification device 302 performs the processing from S802 onward in accordance with the notification from the tracking device 200 indicating whether or not the moving body is being tracked. .

The collation device 302 according to this modification has the same effects as the collation device 300 described in the first embodiment.

Second Embodiment In the second embodiment, when the matching device matches the feature amount of the detected moving object with the feature amount included in the moving object information, the time at which the feature amount is extracted is taken into consideration. An example of

The configuration of the tracking system according to this embodiment is the same as that of the first embodiment. Hereinafter, the description of the contents of the operation of the tracking system according to the present embodiment that overlaps with the description of the first embodiment will be omitted.

Operations of the imaging device 100 and the matching device 300 according to this embodiment will be described.

In the present embodiment, in the process of S706, the imaging device 100 supplies the tracking device 200 and the matching device 300 with the feature amount of the detected moving object, the position information, the area identification information, and the time at which the feature amount was extracted. and send.

In the process of S801, the matching device 300 receives the detected feature amount of the moving object, the position information, the area identification information, and the time when the feature amount was extracted. If the detected moving object is being tracked (“YES” in S803), the matching unit 320 receives the feature amount, the information indicating the boundary area, and the feature amount received from the imaging device 100 in the process of S804. The extracted time (first time) is associated with the identification ID received from the tracking device 200 and stored as mobile body information. FIG. 14 is a diagram showing an example of mobile object information according to the present embodiment.

If the detected moving object is not being tracked ("NO" in S803), in the process of S806, the matching unit 320 extracts the feature amount associated with the corresponding area selected by the area selecting unit 310 from the moving object information. read out. Then, the collating unit 320 collates the detected feature amount of the moving object with the feature amount of the detected moving object whose difference from the time when the feature amount of the detected moving object was extracted is within a predetermined time. do. For example, in the information received by the verification device 300 from the imaging device 100, it is assumed that the feature amount of the detected moving object is "feature amount e" and the time is "12:07". It is also assumed that the areas selected by the area selection unit 310 are "border area x2", "border area x3", and "border area x4". Assuming that the predetermined time is 5 minutes, the feature amount to be compared with the feature amount of the detected moving object among the moving object information shown in FIG. be.

Note that the time used in this embodiment may be the time when the moving object is detected or the time when the imaging data is generated.

Also, in the present embodiment, the predetermined time may differ for each boundary area. For example, when the regions selected by the region selection unit 310 are “boundary region x2”, “boundary region x3” and “boundary region x4”, the predetermined time is set to 5 minutes, and “boundary region x1” and “boundary region x5” are selected. , the predetermined time may be 3 minutes.

As described above, in the second embodiment, in the moving body information, the feature amount of the second moving body is further associated with the time at which the second moving body was present in the boundary area. Then, the matching device 300 according to the second embodiment determines that the difference between the time when the first moving object appears and the first time, among the feature values associated with the selected corresponding region, is a predetermined value. The feature amount associated with the first time, which is within the time period, is collated with the feature amount of the first moving object extracted when the first moving object appears. With this configuration, it is possible to narrow down the objects to be matched with the detected feature amount of the moving object. can be further suppressed.

Further, in the second embodiment, the collation unit 320 may acquire the current time from the system time of the collation device 300 or the like. Then, the matching unit 320 may discard the feature amount associated with the time when the difference from the current time is equal to or greater than a predetermined time, among the mobile object information. Note that the moving body information storage unit 420 may discard the feature amount instead of the matching unit 320 .

In this way, by discarding the feature amount according to the time, it is possible to suppress an increase in the data amount of the mobile body information.

Third Embodiment In the third embodiment, when a moving body that has moved from outside the imaging range to the boundary area moves outside the imaging range using a moving device that allows the moving body to enter, An example in which a verification device performs verification in consideration of operation information of a mobile device will be described.

Specifically, in the situation shown in FIG. 2, it is assumed that the moving object moves from the boundary area x2 to the boundary area x3 using an elevator. At this time, it is assumed that the elevator is restricted to stop only on the first and second floors. In this case, it is not necessary to collate the feature amount of the moving object that has moved to the boundary area x3 with the feature amount of the moving object extracted in the boundary area x4. In addition, in the situation shown in FIG. 2, when the moving object moves from the elevator whose traveling direction is upward to the boundary region x3, it is highly likely that the moving object got on the elevator from the first floor. Even in this case, it is not necessary to collate the feature amount of the moving object that has moved to the boundary area x3 with the feature amount of the moving object extracted in the boundary area x4. In the present embodiment, the verification device performs verification of feature amounts in consideration of the operation information of such moving devices such as elevators.

FIG. 15 is a block diagram illustrating a tracking system 1004 according to a third embodiment of the disclosure. As shown in FIG. 15 , the configuration of a tracking system 1004 according to this embodiment is the same as that of the first embodiment, except that the storage device 401 includes an operating information storage unit 430 . In the following, the description of the configuration and operation of the tracking system 1004 that overlap with the description of the first embodiment will be omitted.

The operating information storage unit 430 stores operating information. FIG. 16 is a diagram illustrating an example of operating information according to the third embodiment; FIG. 16 is operating information about elevators in the situation shown in FIG. In the example of FIG. 16, floors where the elevator stops are determined for each time, but examples of operation information are not limited to this. For example, the information may be information in which the direction in which the elevator travels is associated with each time, or information in which the time when the elevator stops and the floor where the elevator stops may be associated.

The operation of tracking system 1004 will now be described. The area correspondence information storage unit 410 stores the area correspondence information shown in FIG. 5, the moving body information storage unit 420 stores the moving body information shown in FIG. is stored. Also, in this embodiment, the operation of the tracking system 1004 will be described using the situation shown in FIG. 2, but the scene to which the tracking system 1004 is applied is not limited to this example.

In the process of S706, the imaging apparatus 100 transmits the feature amount of the detected moving object, the position information, the area identification information, and the time when the moving object was detected.

In the process of S801, the matching device 300 receives the feature amount of the detected moving object, the position information, the area identification information, and the time when the moving object was detected.

The area selection unit 310 selects the corresponding area and reads the operation information from the operation information storage unit 430 in the process of S805. Based on the read operation information, the area identification information, and the time at which the moving object was detected, the area selecting unit 310 selects a boundary area in which the detected moving object exists among the selected corresponding areas. Select the corresponding area. For example, it is assumed that among the information received by the verification device 300 from the imaging device 100, the information indicating the boundary region included in the region identification information is "boundary region x3" and the time is "9:00". At this time, the area selection unit 310 selects "boundary area x2", "boundary area x3", and "boundary area x4" corresponding to "boundary area x3" from the area correspondence information shown in FIG. Then, since the time is "9:00", the operating state of the elevator is "stopped only on the first and second floors" according to the operating information shown in FIG. Therefore, the region selection unit 310 further selects the “boundary region x2” and the “boundary region x3” from the selected corresponding regions. Then, the collation unit 320 collates the detected moving object feature quantity with the feature quantity associated with the “boundary region x2” or the “boundary region x3” in the moving body information.

In addition, in the present embodiment, the matching device 300 may include the region correspondence information storage unit 410 , the moving object information storage unit 420 , and the operation information storage unit 430 .

Also, examples of mobile devices are not limited to elevators. For example, the mobile device may be an escalator, a train, or the like.

As described above, the verification device 300 according to the third embodiment can detect the first moving object when it appears in one of the plurality of areas from outside the imaging range using the moving device. A corresponding area corresponding to the operation information of the mobile device is selected from the corresponding areas associated with the appearing area. With this configuration, it is possible to narrow down the objects to be matched with the detected feature amount of the moving object. can be further suppressed.

The present disclosure has been described above with reference to the above-described embodiments. However, the disclosure is not limited to the embodiments described above. That is, within the scope of the present disclosure, various aspects that can be understood by those skilled in the art, such as various combinations and selections of the various disclosed elements described above, can be applied.

Some or all of the above-described embodiments can also be described in the following supplementary remarks, but are not limited to the following.

(Appendix 1)
each of a plurality of regions predetermined within an imaging range and a corresponding region that is one of the plurality of regions and is associated with each of the plurality of regions based on region correspondence information; Selecting the corresponding area associated with the area where the first moving object was present,
an area in which a second moving object exists among the plurality of areas, and a feature amount extracted when the second moving object exists in the area, wherein 2, the feature amount associated with the selected corresponding area is acquired from the moving body information including the feature amount of the moving body of No. 2, and when the acquired feature amount and the first moving body exist. Matching with the feature amount of the first moving object extracted in
Matching method.

(Appendix 2)
The collation according to appendix 1, wherein the feature amount of the first moving object is a feature amount extracted when the first moving object moves from outside the imaging range to one of the plurality of regions. Method.

(Appendix 3)
3. The collation method according to any one of appendices 1 and 2, wherein the characteristic amount of the first moving body and the area where the first moving body existed are associated with each other and set in the moving body information.

(Appendix 4)
4. The collation according to any one of appendices 1 to 3, wherein if there is a feature amount that matches the feature amount of the first moving object as a result of the collation, the matching feature amount is discarded from the moving object information. Method.

(Appendix 5)
If the moving object information includes a feature amount of the first moving object that is associated with an area that is not associated with an area where the first moving object exists, 5. The collation method according to any one of Appendices 1 to 4, wherein the feature amount of one moving object is discarded from the moving object information.

(Appendix 6)
wherein the moving object information further associates a first time, which is a time when the second moving object exists in the region, with the characteristic quantity of the second moving object;
a feature amount associated with the first time, which is within a predetermined value from the time at which the first moving object exists, among the feature amounts associated with the selected corresponding region; 6. The collation method according to any one of appendices 1 to 5, wherein the feature amount of the first moving object extracted when the first moving object exists is collated.

(Appendix 7)
When the first moving body moves from outside the imaging range to one of the plurality of areas using a moving device, out of the corresponding areas associated with the area to which the first moving body has moved 7. The collation method according to any one of Appendices 1 to 6, wherein the corresponding area is selected according to operation information of the mobile device.

(Appendix 8)
each of a plurality of regions predetermined within an imaging range and a corresponding region that is one of the plurality of regions and is associated with each of the plurality of regions based on region correspondence information; an area selection means for selecting the corresponding area associated with the area in which the first moving object was present;
an area in which a second moving object exists among the plurality of areas, and a feature amount extracted when the second moving object exists in the area, wherein 2, the feature amount associated with the selected corresponding area is acquired from the moving body information including the feature amount of the moving body of No. 2, and when the acquired feature amount and the first moving body exist. and a matching means for matching the feature amount of the first moving object extracted in the above.

(Appendix 9)
The verification according to appendix 8, wherein the feature amount of the first moving object is a feature amount extracted when the first moving object moves from outside the imaging range to one of the plurality of regions. Device.

(Appendix 10)
10. The method according to any one of appendices 8 and 9, wherein the collating means associates the feature amount of the first moving object with an area in which the first moving object existed, and sets the information in the moving object information. Verification device as described.

(Appendix 11)
11. Any one of appendices 8 to 10, wherein when there is a feature amount that matches the feature amount of the first moving object as a result of the matching, the matching means discards the matching feature amount from the moving object information. 3. A matching device as described in paragraph 1 above.

(Appendix 12)
The collating means may include, in the moving body information, a feature amount of the first moving body that is associated with an area that is not associated with an area in which the first moving body exists. 12. The collation device according to any one of appendices 7 to 11, wherein, if the first moving object is detected, the feature amount of the first moving object is discarded from the moving object information.

(Appendix 13)
wherein the moving object information further associates a first time, which is a time when the second moving object exists in the region, with the characteristic quantity of the second moving object;
The collation means associates the feature value associated with the selected corresponding region with the first time that differs from the time at which the first moving object exists within a predetermined value. 13. The collation device according to any one of appendices 7 to 12, wherein the feature amount of the first moving object is collated with the feature amount of the first moving object extracted when the first moving object exists.

(Appendix 14)
When the first moving object moves from outside the imaging range to one of the plurality of areas using a moving device, the area selecting means associates the moving area with the moving area of the first moving object. 14. The verification device according to any one of appendices 7 to 13, wherein the corresponding region is selected from the corresponding regions corresponding to the movement information of the mobile device.

(Appendix 15)
each of a plurality of regions predetermined within an imaging range and a corresponding region that is one of the plurality of regions and is associated with each of the plurality of regions based on region correspondence information; Among them, a process of selecting the corresponding area associated with the area where the first moving object was present;
an area in which a second moving object exists among the plurality of areas, and a feature amount extracted when the second moving object exists in the area, wherein When the feature amount associated with the selected corresponding region is acquired from the moving body information including the feature amount of one moving body, and the acquired feature amount and the second moving body exist. A program for causing a computer to execute a process of collating with the feature amount of the second moving object extracted in the above.

10 computer device 11 processor 12 RAM
13 ROMs
14 storage device 15 input/output interface 16 bus 17 drive device 18 program 19 peripheral device 20 recording medium 100, 101, 102 imaging device 110 imaging unit 120 moving object detection unit 130 region determination unit 140 feature quantity extraction unit 200 tracking device 300, 301 , 302 verification device 310 region selection unit 320 verification unit 400, 401 storage device 410 region correspondence information storage unit 420 moving object information storage unit 1000, 1001, 1002, 1003, 1004 tracking system

Claims (8)

The first moving object has appeared from the area correspondence information indicating the corresponding area which is at least one of a plurality of boundary areas predetermined within the imaging range and which corresponds to each of the boundary areas. selecting the corresponding region corresponding to the boundary region ;
The selected moving body information in which the feature amount extracted when the second moving body was present in the boundary area and the boundary area in which the second moving body was present are associated with each other . Acquiring a feature amount associated with the corresponding region, comparing the acquired feature amount with the feature amount of the first moving object extracted when the first moving object appears ,
When the first moving body appears in an area movable from the elevator in the boundary area,
In the selecting step, based on the operation information indicating floors where the elevator can stop at each time and the time when the first moving body is detected, at the detected time, the first selecting the corresponding area that can move via an elevator to the boundary area where the moving body of
Matching method. The verification according to claim 1, wherein the feature amount of the first moving object is a feature amount extracted when the first moving object moves from outside the imaging range to one of the boundary areas . Method. 3. The moving object information according to claim 1, wherein the moving object information is updated by associating the characteristic amount of the first moving object with the boundary area where the first moving object appears . Matching method. 4. The method according to any one of claims 1 to 3, wherein if there is a feature amount that matches the feature amount of the first moving object as a result of collation, the matching feature amount is discarded from the moving object information. Matching method. The moving body information includes a feature amount of the first moving body that is associated with the boundary area that is not associated with the boundary area where the first moving body appears . 5. The matching method according to any one of claims 1 to 4, wherein if the first moving object is detected, the feature amount of the first moving object is discarded from the moving object information. wherein the moving object information further associates a time at which the second moving object was present in the boundary area with the characteristic quantity of the second moving object;
Among the feature amounts associated with the selected corresponding region, the feature amount associated with a first time that is a time at which the difference from the time when the first moving body appears is within a predetermined value. 6. The collation method according to any one of claims 1 to 5, wherein the feature amount of said first moving object extracted when said first moving object appears is collated with said first moving object. The first moving object has appeared from the area correspondence information indicating the corresponding area which is at least one of a plurality of boundary areas predetermined within the imaging range and which corresponds to each of the boundary areas. an area selection means for selecting the corresponding area corresponding to the boundary area ;
The selected moving body information in which the feature amount extracted when the second moving body was present in the boundary area and the boundary area in which the second moving body was present are associated with each other . Verification means for obtaining a feature amount associated with the corresponding region and comparing the obtained feature amount with the feature amount of the first moving object extracted when the first moving object appears . and
When the first moving body appears in an area movable from the elevator in the boundary area,
The region selecting means selects the first floor at the detected time based on the operation information indicating floors on which the elevator can stop for each time and the time when the first moving object is detected. selecting the corresponding area that can move via an elevator to the boundary area where the moving body of
Verification device. The first moving object has appeared from the area correspondence information indicating the corresponding area which is at least one of a plurality of boundary areas predetermined within the imaging range and which corresponds to each of the boundary areas. a process of selecting the corresponding area corresponding to the boundary area ;
The selected moving body information in which the feature amount extracted when the second moving body was present in the boundary area and the boundary area in which the second moving body was present are associated with each other . a process of acquiring a feature amount associated with the corresponding region and collating the acquired feature amount with the feature amount of the second moving object extracted when the second moving object appears ; , on the computer , and
When the first moving body appears in an area movable from the elevator in the boundary area,
In the selecting process, based on the operation information indicating floors where the elevator can stop at each time and the time when the first moving body is detected, at the detected time, the first selecting the corresponding area that can move via an elevator to the boundary area where the moving body of
program.

Images