JP7412614B1 - Tracking devices, tracking methods and tracking programs - Google Patents

Abstract

An object of the present invention is to remove appropriate movement trajectories from candidates for association. A tracking unit 25 selects a plurality of image data obtained by photographing a photographing area by a camera 31 as target image data in chronological order, and selects one image data detected from past image data than the target image data. The movement trajectory is updated by associating the movement trajectory of each of the above objects with each of one or more objects detected from the image data of the target. The exclusion unit 26 excludes movement trajectories of one or more objects that have not been associated with the object detected from the image data of the object by the tracking unit 25, and that excludes the shooting area. A movement trajectory in which the position of the object detected from the image data before the image data of the object is included in the exit area through which the object passes before exiting from the included tracking area is excluded from the update target by the tracking unit 25. [Selection diagram] Figure 1

Description

The present disclosure relates to a technique for tracking an object shown in video data.

Updating the movement trajectory by associating M movement trajectories obtained from past image data among the image data constituting the video data with objects detected from new image data. is being carried out.
At this time, as the number M of movement trajectories that are candidates for association with the target object detected from the new image data increases, the possibility that an erroneous association will be made increases. Furthermore, as the number M increases, the amount of calculation involved in the mapping process increases. Therefore, excluding movement trajectories that do not need to be associated from candidates for association improves tracking accuracy and reduces the amount of calculation.

Patent Document 1 describes tracking an object shown in video data obtained by photographing the front of a vehicle. In Patent Document 1, when the current position of the tracked object is within a predetermined area including the edge of the shooting area, if no detected object is associated with the tracked object, it is estimated that the tracked object has moved outside the shooting area. is listed.

Japanese Patent Application Publication No. 2021-117760

In Patent Document 1, since the video data is obtained by photographing the front of the vehicle, it is assumed that the object exits the photographing area only from the edge of the photographing area. However, the object does not necessarily exit the imaging area only from the edge of the imaging area. For example, there are cases where the photographing area is inside a room, and there is a door in the center of the photographing area for exiting from the room. In this case, the object moves out of the imaging area from the door in the center of the imaging area.
Further, even if an object located at the edge of the imaging area is temporarily not detected, this does not necessarily mean that the object can be removed from the tracking target. For example, there may be a case where the photographing area is inside a room, and the edge of the room slightly protrudes from the photographing area. It is assumed that there is no entrance/exit to the room in the part that is out of the shooting area. In this case, even if the object moves out of the imaging area from the edge of the imaging area, the object will always return to the imaging area. Therefore, an object that has moved out of the imaging area from the edge of the imaging area should not be excluded from the tracking target.
The present disclosure aims to make it possible to exclude appropriate movement trajectories from candidates to be associated.

The tracking device according to the present disclosure includes:
A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking unit that updates the movement trajectory by associating each of the objects with one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that is not correlated with the object detected from the image data of the object by the tracking unit, and includes the photographed area. an exclusion unit that excludes a movement trajectory that includes a position of the object detected from image data preceding the image data of the object in an exit area through which the object passes before exiting the area from an update target by the tracking unit; Be prepared.

In the present disclosure, a movement trajectory in which the position of the object detected from the image data before the image data of the object is included in the exit area through which the object passes before exiting from the tracking area is excluded from the update target. This makes it possible to remove appropriate movement trajectories from the candidates to be associated.

1 is a configuration diagram of a tracking device 10 according to Embodiment 1. FIG. FIG. 3 is an explanatory diagram of area setting processing according to the first embodiment. 5 is a flowchart of tracking processing according to the first embodiment. FIG. 3 is an explanatory diagram of first exclusion processing according to the first embodiment. FIG. 7 is an explanatory diagram of second exclusion processing according to the first embodiment. 7 is a flowchart of tracking processing according to modification 1. FIG. 3 is a configuration diagram of a tracking device 10 according to a second modification. 7 is a flowchart of tracking processing in a set period according to Embodiment 2. FIG. 7 is an explanatory diagram of the results of tracking processing in a set period according to the second embodiment. FIG. 7 is an explanatory diagram of method 1 according to Embodiment 2. FIG. 7 is an explanatory diagram of method 2 according to the second embodiment. FIG. 7 is an explanatory diagram of a method of enclosing a hit point according to Embodiment 2;

Embodiment 1.
***Explanation of configuration***
The configuration of tracking device 10 according to Embodiment 1 will be described with reference to FIG. 1.
Tracking device 10 is a computer.
The tracking device 10 includes hardware of a processor 11, a memory 12, a storage 13, and a communication interface 14. The processor 11 is connected to other hardware via signal lines and controls these other hardware.

The processor 11 is an IC that performs processing. IC is an abbreviation for Integrated Circuit. Specific examples of the processor 11 include a CPU, a DSP, and a GPU. CPU is an abbreviation for Central Processing Unit. DSP is an abbreviation for Digital Signal Processor. GPU is an abbreviation for Graphics Processing Unit.

The memory 12 is a storage device that temporarily stores data. The memory 12 is, for example, an SRAM or a DRAM. SRAM is an abbreviation for Static Random Access Memory. DRAM is an abbreviation for Dynamic Random Access Memory.

The storage 13 is a storage device that stores data. The storage 13 is, for example, an HDD. HDD is an abbreviation for Hard Disk Drive. Furthermore, the storage 13 may be a portable recording medium such as an SD (registered trademark) memory card, CompactFlash (registered trademark), NAND flash, flexible disk, optical disk, compact disc, Blu-ray (registered trademark) disk, or DVD. good. SD is an abbreviation for Secure Digital. DVD is an abbreviation for Digital Versatile Disk.

The communication interface 14 is an interface for communicating with an external device. The communication interface 14 is, for example, an Ethernet (registered trademark), USB, or HDMI (registered trademark) port. USB is an abbreviation for Universal Serial Bus. HDMI is an abbreviation for High-Definition Multimedia Interface.

Tracking device 10 is connected to camera 31 via communication interface 14 .

The tracking device 10 includes a region setting section 21, an image acquisition section 22, a detection section 23, a feature calculation section 24, a tracking section 25, and an exclusion section 26 as functional components. The functions of each functional component of the tracking device 10 are realized by software.
The storage 13 stores programs that implement the functions of each functional component of the tracking device 10. This program is read into the memory 12 by the processor 11 and executed by the processor 11. Thereby, the functions of each functional component of the tracking device 10 are realized.

Furthermore, an exit area 41 is stored in the storage 13 .

In FIG. 1, only one processor 11 was shown. However, there may be a plurality of processors 11, and the plurality of processors 11 may cooperate to execute programs that implement each function.

***Operation explanation***
The operation of the tracking device 10 according to the first embodiment will be described with reference to FIGS. 2 to 5.
The operation procedure of the tracking device 10 according to the first embodiment corresponds to the tracking method according to the first embodiment. Furthermore, a program that realizes the operation of the tracking device 10 according to the first embodiment corresponds to the tracking program according to the first embodiment.

The operation of the tracking device 10 includes area setting processing and tracking processing. The area setting process is a process for setting the exit area 41. The tracking process is a process of tracking a target object. In the first embodiment, it is assumed that the object is a person. However, the target object is not limited to a person, but may be another type of object such as a car or a bag.

Region setting processing according to the first embodiment will be described with reference to FIG. 2.
The area setting unit 21 sets the exit area 41 according to the administrator's operation. The exit area 41 is written to the storage 13. The exit area 41 is an area through which the object passes before exiting the tracking area 42 . The tracking area 42 is an area for tracking an object, and is an area that includes the photographing area 32 of the camera 31. The exit area 41 may have any shape and may be set in a plurality of locations.
In the photographing area 32 shown in FIG. 2, since the person as the object moves through the floor, only the periphery of the end of the floor portion among the ends of the photographing area 32 is set as the exit area 41. Further, an exit area 41 is also set on the floor in front of the door. When the camera 31 is a surveillance camera or the like, the exit area 41 is defined as at least the peripheral area of a part of the end of the photographing area 32 and the area other than the end of the photographing area, such as in front of the door. It is assumed that either of these will occur.

With reference to FIG. 3, tracking processing according to the first embodiment will be described.
(Step S11: Image acquisition processing)
The image acquisition unit 22 acquires a plurality of image data forming the video data obtained by photographing the photographing area 32 by the camera 31 as target image data in chronological order.

(Step S12: Detection process)
The detection unit 23 detects one or more target objects from the target image data acquired in step S11.
Specifically, the detection unit 23 detects the target object by inputting the image data of the target to an object detection model that detects the target object from image data, which is configured using a neural network or the like. . Here, it is assumed that a rectangular detection frame surrounding the detected object is obtained.

The processes in step S13 and step S14 are executed using each object detected in step S12 as a process target.

(Step S13: Feature amount calculation process)
The feature amount calculation unit 24 calculates the feature amount of the object. Existing techniques may be used as a method for calculating feature amounts.

(Step S14: Correlation processing)
The tracking unit 25 associates the movement trajectory of each of one or more objects detected from image data older than the image data of the target with the object to be processed detected from the image data of the target. conduct. Thereby, the tracking unit 25 updates the movement trajectory.
Note that the movement trajectory that is a candidate to be associated here is a movement trajectory that has not been removed from the update target in the process of step S15 or step S16, which will be described later.

Specifically, the tracking unit 25 executes the processes from step S141 to step S143.
(Step S141: Dissimilarity calculation process)
The tracking unit 25 sets each movement trajectory that is not excluded from the update target as the target movement trajectory. The tracking unit 25 calculates the degree of difference between the recent object in the movement trajectory of the object and the object to be processed. The most recent object in the movement trajectory of the target is the object that was added last to the movement trajectory of the target in the process of step S142 or step S143, which will be described later.
Specifically, the tracking unit 25 calculates a first distance, which is the distance between the most recent position of the object and the position of the object to be processed, and the feature amount of the most recent object and the position of the object to be processed. The degree of dissimilarity is calculated from at least one of the second distance, which is the distance between the feature amount and the second distance. For example, the tracking unit 25 calculates the weighted sum of the first distance and the second distance as the degree of dissimilarity.
In the first embodiment, since the target object is a person who is required to be in contact with the ground, the position of the target object is the position of the center of the lower side of the detection frame obtained in step S12. Depending on the target object, the position of the target object may be the center of the detection frame obtained in step S12. Euclidean distance can be considered as the distance between positions. A cosine distance can be considered as the distance between the feature amounts.

The tracking unit 25 advances the process to step S142 if there is a movement trajectory with a degree of difference equal to or less than the threshold value. On the other hand, if there is no movement trajectory with a degree of difference equal to or less than the threshold value, the tracking unit 25 advances the process to step S143.

(Step S142: Additional processing)
The tracking unit 25 adds the object to be processed to the movement trajectory with the smallest degree of difference calculated in step S141.

(Step S143: New trajectory generation process)
The tracking unit 25 generates a new movement trajectory and adds an object to be processed.

The processes in step S15 and step S16 are executed for each movement trajectory to which no target object was added in step S14.

(Step S15: First exclusion process)
The exclusion unit 26 excludes the object from being updated by the tracking unit 25 when the movement trajectory of the object is not correlated with the object detected from the reference number of chronologically continuous image data.

This will be explained in detail with reference to FIG. 4.
In FIG. 4, the reference number is 3. Further, the target image data is assumed to be image data t. Since the reference number is 3, the exclusion unit 26 excludes movement trajectories that have not been associated with objects detected from three consecutive image data including the target image data from being updated by the tracking unit 25. remove. Therefore, for all three image data t, image data t-1, and image data t-2, movement trajectories that are not associated with detected objects are excluded from the update target.
The movement trajectory A is associated with an object detected from image data t, which is image data of the object. Therefore, the movement trajectory A is not subject to the processing in step S15 in the first place, and is not excluded from the update subject.
The movement trajectory B is not associated with the object detected from the image data t and image data t-1. However, the movement trajectory B is associated with the object detected from the image data t-2. Therefore, movement trajectory B is not excluded from the update target.
The movement trajectory C is not associated with any detected object for all three of the image data t, image data t-1, and image data t-2. Therefore, the movement trajectory C is excluded from the update target.

(Step S16: Second exclusion process)
The exclusion unit 26 excludes the object from being updated by the tracking unit 25 when the movement trajectory of the object includes the position of the object detected from the image data before the image data of the object in the exit area 41. The position of the object detected from the image data before the image data of the object is the closest position of the movement trajectory of the object. That is, a movement trajectory that is not associated with a target object detected from the target image data and whose most recent position is in the exit area 41 is excluded from the update target. This means that if the object is not associated with the object detected from the image data of the target and its most recent position is in the exit area 41, the object indicated by the movement trajectory is presumed to have gone outside the tracking area 42. This is to be done.

This will be explained in detail with reference to FIG.
In FIG. 5, the exit area 41 is set as shown in FIG.
Regarding the movement trajectory D, the position of the object detected from the image data before the image data of the object is included in the exit area 41. However, the movement trajectory D is associated with the target object detected from the image data t, which is the image data of the target. Therefore, the movement trajectory D is not subject to the processing in step S16 in the first place, and is not excluded from the update subject.
Regarding the movement trajectory E, the position of the object detected from the image data before the image data of the object is not included in the exit area 41. Therefore, the movement trajectory E is not excluded from the update target.
Regarding the movement trajectory F, the position of the object detected from the image data before the image data of the object is included in the exit area 41. Therefore, the movement trajectory F is excluded from the update target.

The process returns to step S11, and the next image data in time order is acquired as target image data.

***Effects of Embodiment 1***
As described above, in the tracking device 10 according to the first embodiment, the position of the object detected from the image data before the image data of the object is in the exit area 43 through which the object passes before exiting from the tracking area 42. Remove the included movement trajectory from the update target.
Thereby, a movement trajectory in which the object indicated by the movement trajectory is estimated to have moved outside the tracking area 42 can be excluded from the update target. Therefore, it becomes possible to remove appropriate movement trajectories from the candidates to be associated. By excluding appropriate movement trajectories from candidates to be associated, it is possible to improve tracking accuracy and reduce the amount of calculation.

***Other configurations***
<Modification 1>
As a first modification, the tracking unit 25 may associate objects with the same attribute with each other, and may not associate objects with different attributes with each other. For example, assume that the objects are people and vehicles. In this case, only people are associated with a person's movement trajectory, and only vehicles are associated with a vehicle movement trajectory.

With reference to FIG. 6, tracking processing according to modification 1 will be described.
The processing in step S13 and step S14 is executed for each attribute, and for each object of the target attribute detected in step S12. As a result, in step S14, the tracking unit 25 associates objects with the same attribute with each other, and does not associate objects with different attributes with each other.

This makes it possible to suppress the possibility of erroneously tracking different objects as the same object.

<Modification 2>
In the first embodiment, each functional component is realized by software. However, as a second modification, each functional component may be realized by hardware. Regarding this second modification, the differences from the first embodiment will be explained.

With reference to FIG. 7, the configuration of the tracking device 10 according to Modification 2 will be described.
When each functional component is realized by hardware, the tracking device 10 includes an electronic circuit 15 instead of the processor 11, memory 12, and storage 13. The electronic circuit 15 is a dedicated circuit that realizes the functions of each functional component, the memory 12, and the storage 13.

The electronic circuit 15 may be a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA, an ASIC, or an FPGA. GA is an abbreviation for Gate Array. ASIC is an abbreviation for Application Specific Integrated Circuit. FPGA is an abbreviation for Field-Programmable Gate Array.
Each functional component may be realized by one electronic circuit 15, or each functional component may be realized by being distributed among a plurality of electronic circuits 15.

<Modification 3>
As a third modification, some of the functional components may be realized by hardware, and other functional components may be realized by software.

The processor 11, memory 12, storage 13, and electronic circuit 15 are referred to as a processing circuit. That is, the functions of each functional component are realized by the processing circuit.

Embodiment 2.
The second embodiment differs from the first embodiment in that the exit area 41 is set without manual intervention. In the second embodiment, this different point will be explained, and the explanation of the same point will be omitted.

***Operation explanation***
The operation of the tracking device 10 according to the second embodiment will be described with reference to FIGS. 8 to 12.
The operation procedure of the tracking device 10 according to the second embodiment corresponds to the tracking method according to the second embodiment. Further, a program that realizes the operation of the tracking device 10 according to the second embodiment corresponds to the tracking program according to the second embodiment.

Method 1 and Method 2 will be explained as methods for setting the exit area 41.
In method 1 and method 2, first, as shown in FIG. 8, during a set period, the tracking process shown in FIG. 3 except for step S16 is executed. The setting period is a preparation period before starting the tracking process. Then, as shown in FIG. 9, a plurality of movement trajectories are identified. Each movement trajectory starts from the target object that was added when it was newly created in step S143, and continues to the target object that was added immediately before being excluded from the update target in step S15.

<Method 1>
Method 1 will be explained with reference to FIG.
The area setting unit 21 places a point at the first position in each movement trajectory generated from the image data obtained by the camera 31 during the setting period. That is, the area setting unit 21 places a dot on each movement locus at the position of the object added when newly created in step S143. This position is estimated to indicate the position immediately after the object enters the photographing area 32. Note that although the number of dots is small in FIG. 10, in reality, by setting a period of a certain length as the set period, many dots can be hit. The area setting unit 21 sets the outer edge of the area surrounding the hit point as the exit area 41. A convex hull may be used as a method for specifying the area surrounding the hit point.
The place where the object entered is estimated to be the place through which the object passed before leaving. Therefore, by setting the outer edge of the area surrounding the position where the object entered as the exit area 41, the exit area 41 can be appropriately set.

<Method 2>
Method 2 will be explained with reference to FIG.
The area setting unit 21 places a dot at the last position in each movement trajectory generated from the image data obtained by the camera 31 during the setting period. That is, the area setting unit 21 places a dot on each movement locus at the position of the object that was added immediately before being excluded from the update target in step S15. This position is estimated to indicate the position immediately before the object leaves the imaging area 32. Note that although the number of dots is small in FIG. 11, in reality, by setting a period of a certain length as the set period, many dots can be hit. The area setting unit 21 sets the outer edge of the area surrounding the hit point as the exit area 41.
The location immediately before the object exits is estimated to be the location the object passes through before exiting. Therefore, by setting the outer edge of the area surrounding the position immediately before the object exits as the exit area 41, the exit area 41 can be appropriately set.

Note that by combining method 1 and method 2, the area setting unit 21 sets points at both the first position and the last position in each movement trajectory generated from image data obtained by the camera 31 during the setting period. You can hit it.

In method 1 and method 2, when enclosing the hit point, the area setting unit 21 may perform clustering for each adjacent point and surround the point for each cluster. Then, the area setting unit 21 may set the exit area 41 as the inside of the enclosed range or the outer edge of the enclosed range. For example, as shown in FIG. 12(A), when all points are surrounded at once, there is a possibility that exit areas 41 may be set in areas other than the locations where the object passes before exiting. On the other hand, as shown in FIG. 12(B), by clustering, it is possible to prevent the exit area 41 from being set at a location other than the location where the object passes before exiting.

***Effects of Embodiment 2***
As described above, the tracking device 10 according to the second embodiment sets the exit area 41 without manual intervention. This eliminates the need to set the exit area 41 for each installation position of the camera 31, and the operational load on the tracking device 10 can be reduced.

Note that the area setting unit 21 may accept corrections to the exit area 41 set by method 1 or method 2. For example, it is conceivable that the administrator looks at the exit area 41 set in method 1 or method 2 and adjusts the shape.

"Unit" in the above description may be read as "circuit," "step," "procedure," "process," or "processing circuit."

Hereinafter, various aspects of the present disclosure will be collectively described as supplementary notes.
(Additional note 1)
A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking unit that updates the movement trajectory by associating each of the objects with one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that is not correlated with the object detected from the image data of the object by the tracking unit, and includes the photographed area. an exclusion unit that excludes a movement trajectory that includes a position of the object detected from image data preceding the image data of the object in an exit area through which the object passes before exiting the area from an update target by the tracking unit; tracking device.
(Additional note 2)
The tracking device according to appendix 1, wherein the exit area is at least one of a peripheral area of a part of the end of the imaging area and an area other than the end of the imaging area.
(Additional note 3)
The tracking device further includes:
The tracking device according to supplementary note 1, further comprising an area setting unit that sets an outer edge of an area surrounding a point indicating a first position on a movement trajectory generated from image data obtained by the camera during a setting period as the exit area.
(Additional note 4)
The tracking device further includes:
The tracking device according to supplementary note 1, further comprising an area setting unit that sets an outer edge of an area surrounding a point indicating a last position on a movement trajectory generated from image data obtained by the camera during a setting period as the exit area.
(Appendix 5)
The exclusion unit excludes a movement trajectory that has not been associated with an object detected from a reference number of chronologically continuous image data from being updated by the tracking unit. The tracking device described in .
(Appendix 6)
The tracking unit uses a distance between a position of the target detected from the image data of the target and a position of the target detected from image data before the image data of the target on the movement trajectory. , the tracking device according to any one of Supplementary Notes 1 to 5, which correlates the movement trajectory with a target object detected from image data of the target.
(Appendix 7)
The tracking unit calculates the distance between the feature amount of the object detected from the image data of the object and the feature amount of the object detected from the image data before the image data of the object in the movement trajectory. The tracking device according to any one of Supplementary Notes 1 to 6, wherein the tracking device uses the tracking device to associate the movement trajectory with a target object detected from image data of the target.
(Appendix 8)
The tracking device according to any one of Supplementary Notes 1 to 7, wherein the tracking unit associates objects with the same attribute with each other.
(Appendix 9)
A computer uses a plurality of image data obtained by photographing a photographing area with a camera as target image data in chronological order, and for each of one or more target objects detected from image data older than the target image data. updating the movement trajectory by associating the movement trajectory with each of one or more objects detected from the image data of the target;
Among the movement trajectories of each of the one or more objects, the computer determines a tracking area that includes the photographed area, which is a movement trajectory that has not been correlated with the object detected from the image data of the object. A tracking method that excludes from updating a movement trajectory in which an exit area through which the object passes before exiting from the object includes a position of the object detected from image data before image data of the object.
(Appendix 10)
A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking process that updates the movement trajectory by associating the object with each of the one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that is not correlated with the object detected from the image data of the object by the tracking process, and includes the photographed area. Exclusion processing that excludes a movement trajectory in which the position of the object detected from the image data before the image data of the object is included in an exit area through which the object passes before exiting the area from the update target by the tracking process. A tracking program that allows your computer to act as a tracking device.

The embodiments and modifications of the present disclosure have been described above. Some of these embodiments and modifications may be implemented in combination. Moreover, any one or some of them may be partially implemented. Note that the present disclosure is not limited to the above embodiments and modifications, and various changes can be made as necessary.

10 tracking device, 11 processor, 12 memory, 13 storage, 14 communication interface, 15 electronic circuit, 21 area setting unit, 22 image acquisition unit, 23 detection unit, 24 feature amount calculation unit, 25 tracking unit, 26 exclusion unit, 31 Camera, 32 Photographing area, 41 Exit area, 42 Tracking area.

Claims (13)

A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking unit that updates the movement trajectory by associating each of the objects with one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that has not been correlated with the object detected from the image data of the object by the tracking unit, and which is an edge of the photographing area The image data of the object is at least one of a peripheral area of a part of the area and an area other than the edge of the imaging area, and the image data of the object is in an exit area through which the object passes before exiting from a tracking area including the imaging area. A tracking device comprising: an exclusion unit that excludes a movement trajectory including a position of an object detected from previous image data from being updated by the tracking unit. A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking unit that updates the movement trajectory by associating each of the objects with one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that is not correlated with the object detected from the image data of the object by the tracking unit, and includes the photographed area. an exclusion unit that excludes a movement trajectory that includes a position of the object detected from image data before the image data of the object in an exit area through which the object passes before exiting the area from an update target by the tracking unit ;
an area setting unit that sets an outer edge of an area surrounding a point indicating a first position on a movement trajectory generated from image data obtained by the camera during a setting period as the exit area;
A tracking device comprising: A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking unit that updates the movement trajectory by associating each of the objects with one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that is not correlated with the object detected from the image data of the object by the tracking unit, and includes the photographed area. an exclusion unit that excludes a movement trajectory that includes a position of the object detected from image data before the image data of the object in an exit area through which the object passes before exiting the area from an update target by the tracking unit ;
an area setting unit that sets an outer edge of an area surrounding a point indicating a last position on a movement trajectory generated from image data obtained by the camera during a setting period as the exit area;
A tracking device comprising: Any one of claims 1 to 3, wherein the exclusion unit excludes movement trajectories that have not been associated with objects detected from a reference number of chronologically continuous image data from being updated by the tracking unit. Tracking devices as described in Section . The tracking unit uses a distance between a position of the target detected from the image data of the target and a position of the target detected from image data before the image data of the target on the movement trajectory. The tracking device according to any one of claims 1 to 3, wherein the tracking device associates the movement trajectory with a target object detected from image data of the target. The tracking unit calculates the distance between the feature amount of the object detected from the image data of the object and the feature amount of the object detected from the image data before the image data of the object in the movement trajectory. The tracking device according to any one of claims 1 to 3, wherein the tracking device associates the movement trajectory with a target object detected from image data of the target. The tracking device according to any one of claims 1 to 3, wherein the tracking unit associates objects with the same attribute. A computer uses a plurality of image data obtained by photographing a photographing area with a camera as target image data in chronological order, and for each of one or more target objects detected from image data older than the target image data. updating the movement trajectory by associating the movement trajectory with each of one or more objects detected from the image data of the target;
Of the movement trajectories of each of the one or more objects, the computer detects movement trajectories that are not correlated with the object detected from the image data of the object, and which are located at the edge of the photographing area. At least one of a peripheral region and a region other than the edge of the photographing region, and the front of the image data of the object is an exit region through which the object passes before exiting from a tracking region including the photographing region. A tracking method that excludes movement trajectories that include the position of objects detected from image data from being updated. A computer uses a plurality of image data obtained by photographing a photographing area with a camera as target image data in chronological order, and for each of one or more target objects detected from image data older than the target image data. updating the movement trajectory by associating the movement trajectory with each of one or more objects detected from the image data of the target;
Among the movement trajectories of each of the one or more objects, the computer determines a tracking area that includes the photographed area, which is a movement trajectory that has not been correlated with the object detected from the image data of the object. excluding from update targets a movement trajectory in which an exit area through which the object passes before exiting includes a position of the object detected from image data before the image data of the object;
A tracking method , wherein a computer sets the outer edge of a region surrounding a point indicating a first position on a movement trajectory generated from image data obtained by the camera during a set period as the exit region. A computer uses a plurality of image data obtained by photographing a photographing area with a camera as target image data in chronological order, and for each of one or more target objects detected from image data older than the target image data. updating the movement trajectory by associating the movement trajectory with each of one or more objects detected from the image data of the target;
Among the movement trajectories of each of the one or more objects, the computer determines a tracking area that includes the photographed area, which is a movement trajectory that has not been correlated with the object detected from the image data of the object. excluding from update targets a movement trajectory in which an exit area through which the object passes before exiting includes a position of the object detected from image data before the image data of the object;
A tracking method, wherein a computer sets, as the exit area, an outer edge of an area surrounding a point indicating a last position on a movement trajectory generated from image data obtained by the camera during a set period. A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking process that updates the movement trajectory by associating the object with each of the one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that has not been correlated with the object detected from the image data of the object by the tracking process, and which is an edge of the photographing area. The image data of the object is at least one of a peripheral area of a part of the area and an area other than the edge of the imaging area, and the image data of the object is in an exit area through which the object passes before exiting from a tracking area including the imaging area. A tracking program that causes a computer to function as a tracking device that performs an exclusion process that excludes a movement trajectory that includes a position of an object detected from previous image data from being updated by the tracking process. A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking process that updates the movement trajectory by associating the object with each of the one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that is not correlated with the object detected from the image data of the object by the tracking process, and includes the photographed area. Excluding a movement trajectory in which the position of the object detected from the image data before the image data of the object is included in an exit area through which the object passes before exiting the area from the update target by the tracking process ;
an area setting process of setting an outer edge of an area surrounding a point indicating a first position on a movement trajectory generated from image data obtained by the camera during a setting period as the exit area;
A tracking program that allows a computer to act as a tracking device. A plurality of image data obtained by photographing a photographing area by a camera are set as target image data in time order, and a movement trajectory of each of one or more targets detected from image data older than the target image data. and a tracking process that updates the movement trajectory by associating the object with each of the one or more objects detected from the image data of the object;
Among the movement trajectories of each of the one or more objects, a movement trajectory that is not correlated with the object detected from the image data of the object by the tracking process, and includes the photographed area. Excluding a movement trajectory in which the position of the object detected from the image data before the image data of the object is included in an exit area through which the object passes before exiting the area from the update target by the tracking process ;
an area setting process of setting an outer edge of an area surrounding a point indicating a last position on a movement trajectory generated from image data obtained by the camera during a setting period as the exit area;
A tracking program that allows a computer to act as a tracking device.

Images