JP2021044726A - Notification determination device, notification determination method, and notification determination program - Google Patents

Abstract

To make it possible to appropriately detect people who are swaying from captured image data obtained by a photographing device.SOLUTION: An approximation unit 24 calculates an approximate straight line that approximates a movement locus obtained by connecting, in time series, a plurality of detection points representing positions of a subject detected from captured image data obtained in a reference period. A determination unit 25 determines whether a person is swaying and determines whether to notify based on average distance, which is an average value of distance between the approximate straight line calculated by the approximation unit 24 and each of the detection points, and the number of intersections between the approximate straight line and the movement locus.SELECTED DRAWING: Figure 1

Description

The present invention relates to a technique for detecting and notifying wobble.

There is a technology that detects a person who takes a specific action from the photographed image data acquired by the photographing device and notifies it as an object to be noted. A particular action is wobble, which means meandering and moving.
In Patent Document 1, the distance between the center line connecting the start point of movement and the nth point from the start point and each point is defined as the swing width, and when the swing width is larger than the threshold value, it fluctuates. It is stated that it is judged to be present.

Japanese Unexamined Patent Publication No. 2007-089105

In the technique described in Patent Document 1, there is a possibility that even if it does not actually wobble, such as when it moves in an arc-shaped locus to avoid other people, it may be detected as wobbling.
An object of the present invention is to make it possible to detect a person who wanders appropriately.

The notification determination device according to the present invention is
An approximation part that calculates an approximate straight line that approximates the movement locus obtained by connecting multiple detection points representing the position of the subject detected in the reference period in time series, and
Whether to notify based on the average distance which is the average value of the distances between the approximate straight line and each of the plurality of detection points calculated by the approximate unit and the number of intersections between the approximate straight line and the movement locus. It is provided with a determination unit for determining whether or not.

The determination unit determines to notify when the average distance is equal to or greater than the first threshold value and the number of intersections is equal to or greater than the second threshold value.

The determination unit has the intersection at a moving radius that is the radius of the smallest circle including the plurality of detection points, the average distance of which is equal to or greater than the first threshold value and the number of intersections of which is equal to or greater than the second threshold value. When the crossing ratio divided by the number of times is equal to or greater than the third threshold value, it is determined to notify.

The notification determination device further
A detection unit that detects the target person from the target photographed image data for each of the photographed image data of a plurality of frames obtained by photographing the photographing area with the photographing device.
It is provided with a detection point setting unit that sets a set of points representing the positions of the target person detected by the detection unit from each of the captured image data of the plurality of frames as the plurality of detection points.

The detection point setting unit obtains a plurality of points obtained by excluding points that are separated from a set of points representing the position of the target person by a distance equal to or greater than a limit distance from the previous points in the time series. It is set as the plurality of detection points.

When the movement locus indicates that the determination unit has moved in one direction in the depth direction of the captured image data and then has moved a distance of half or more of the distance moved in the one direction in the opposite direction of the depth direction. Is determined not to be notified.

In the determination unit, the number of detection points between the first intersection where the movement locus intersects the approximate straight line and the second intersection where the movement locus intersects the approximate straight line is the reference number regarding the number of detection points. If it is less, the second intersection is not counted as the number of intersections.

The detection point setting unit corrects the lens distortion of the photographing device for each of the photographed image data of the plurality of frames, converts the bird's-eye view, and generates the bird's-eye view image data. A set of points representing the position of the target person in the bird's-eye view image data is set as the plurality of detection points.

The notification determination method according to the present invention is
The approximation part calculates an approximate straight line that approximates the movement locus obtained by connecting a plurality of detection points representing the positions of the subjects detected in the reference period in chronological order.
The determination unit determines whether or not to notify based on the average distance which is the average value of the distances between the approximate straight line and each of the plurality of detection points and the number of intersections between the approximate straight line and the movement locus. To do.

The notification determination program according to the present invention is
An approximation part that calculates an approximate straight line that approximates the movement locus obtained by connecting multiple detection points representing the position of the subject detected in the reference period in time series, and
Whether to notify based on the average distance which is the average value of the distances between the approximate straight line and each of the plurality of detection points calculated by the approximate unit and the number of intersections between the approximate straight line and the movement locus. The computer functions as a notification determination device that performs a determination unit that determines whether or not.

In the present invention, it is determined whether or not to notify based on the average distance which is the average value of the distances between the approximate straight line and each of the plurality of detection points and the number of intersections between the approximate straight line and the movement locus. This makes it possible to detect wobbling appropriately.

The block diagram of the notification determination apparatus 10 which concerns on Embodiment 1. FIG. The flowchart which shows the process flow of the notification determination apparatus 10 which concerns on Embodiment 1. FIG. The explanatory view of the detection process which concerns on Embodiment 1. FIG. The explanatory view of the grouping process which concerns on Embodiment 1. FIG. An explanatory diagram of data stored in the position storage unit 31 according to the first embodiment. The explanatory view of the approximation process which concerns on Embodiment 1. FIG. The flowchart of the determination process which concerns on Embodiment 1. The explanatory view of the distance between the approximate straight line and the detection point which concerns on Embodiment 1. FIG. The block diagram of the notification determination apparatus 10 which concerns on modification 1. FIG. The flowchart of the determination process which concerns on Embodiment 2. Explanatory drawing of the moving radius which concerns on Embodiment 2. An explanatory diagram of the effect of the notification determination device 10 according to the second embodiment. Explanatory drawing of method 1 which concerns on embodiment 3. FIG. Explanatory drawing of method 2 which concerns on embodiment 3. FIG. Explanatory drawing of method 3 which concerns on embodiment 3. FIG. The flowchart which shows the process flow of the notification determination apparatus 10 which concerns on Embodiment 4. The explanatory view of the process which corrects the lens distortion of the photographing apparatus 41 which concerns on Embodiment 4. FIG. Explanatory drawing of the bird's-eye view conversion which concerns on Embodiment 4.

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

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

The memory 12 is a storage device that temporarily stores data. Specific examples of the memory 12 are SRAM (Static Random Access Memory) and DRAM (Dynamic Random Access Memory).

The storage 13 is a storage device for storing data. As a specific example, the storage 13 is an HDD (Hard Disk Drive). The storage 13 includes SD (registered trademark, Secure Digital) memory card, CF (Compact Flash, registered trademark), NAND flash, flexible disk, optical disk, compact disk, Blu-ray (registered trademark) disk, DVD (Digital Versaille Disk), and the like. It may be a portable recording medium.

The communication interface 14 is an interface for communicating with an external device. Specific examples of the communication interface 14 are Ethernet (registered trademark), USB (Universal Serial Bus), and HDMI (registered trademark, High-Definition Multimedia Interface) ports.

The notification determination device 10 is connected to a photographing device 41 such as a surveillance camera via a communication interface 14.

The notification determination device 10 includes an image acquisition unit 21, a detection unit 22, a detection point setting unit 23, an approximation unit 24, a determination unit 25, and a notification unit 26 as functional components. The functions of each functional component of the notification determination device 10 are realized by software.
The storage 13 stores a program that realizes the functions of each functional component of the notification determination device 10. This program is read into the memory 12 by the processor 11 and executed by the processor 11. As a result, the functions of each functional component of the notification determination device 10 are realized.

Further, the storage 13 realizes the position storage unit 31.

In FIG. 1, only one processor 11 was shown. However, the number of processors 11 may be plural, and the plurality of processors 11 may execute programs that realize each function in cooperation with each other.

*** Explanation of operation ***
The operation of the notification determination device 10 according to the first embodiment will be described with reference to FIGS. 2 to 8.
The operation procedure of the notification determination device 10 according to the first embodiment corresponds to the notification determination method according to the first embodiment. Further, the program that realizes the operation of the notification determination device 10 according to the first embodiment corresponds to the notification determination program according to the first embodiment.

The processing flow of the notification determination device 10 according to the first embodiment will be described with reference to FIG.
(Step S11: Image acquisition process)
The image acquisition unit 21 acquires the latest frame captured image data obtained by photographing the photographing area with the photographing device 41 via the communication interface 14. The image acquisition unit 21 writes the captured image data into the memory 12.

(Step S12: Detection process)
The detection unit 22 reads the captured image data acquired in step S11 from the memory 12. The detection unit 22 detects a person from the target image data.
As shown in FIG. 3, the detection unit 22 cuts out the partial image data of the rectangular region where the detected person is located from the captured image data as the partial image data about the person. The detection unit 22 specifies the coordinates of the center point of the lower side of the cut out partial image data as the position where the person is. That is, the detection unit 22 specifies the coordinates of the person's feet as the position where the person is.

(Step S13: Grouping process)
The detection unit 22 identifies the same person as the person detected in step S12 from the person detected from the past captured image data.
Specifically, as shown in FIG. 4, the detection unit 22 describes the partial image data of the person detected from the photographed image data acquired in step S11 and the person detected from the past photographed image data. Compare with the past data which is the partial image data, and identify the past data which is more similar than the reference degree. The detection unit 22 identifies a person indicated by past data similar to the reference degree or higher as the same person as the detected person.
The detection unit 22 attaches the same identifier as the person identified as the same person to the coordinates and partial image data indicating the position of the detected person, and writes them in the position storage unit 31. When there is no past data similar to the reference degree or more, the detection unit 22 adds a new identifier to the coordinates and partial image data indicating the position of the detected person, and writes them in the position storage unit 31.
By repeatedly executing the process shown in FIG. 2, as shown in FIG. 5, data indicating the history of the position of each person is stored in the position storage unit 31 together with the order of each identifier. As an item for which the order can be understood, the partial image data may be stored together with the acquired time data.

The processes of steps S14 to S16 are executed for each person. That is, the processes of steps S14 to S16 are executed for each identifier stored in the position storage unit 31 within the past reference period. However, identifiers in which the number of records stored in the position storage unit 31 within the past reference period is less than the reference number regarding the number of stored records are excluded. The reference number regarding the number of stored records is an integer of 3 or more. It is desirable that the reference number regarding the number of stored records is a certain number such as 10 or more in order to appropriately detect the wobble.

(Step S14: Detection point setting process)
The detection point setting unit 23 reads out the coordinates from the plurality of records to which the target identifier is attached from the position storage unit 31. The detection point setting unit 23 sets a set of points indicated by the read coordinates, that is, a set of points representing the position of the target person as a plurality of detection points representing the position of the target person detected in the reference period. The detection point setting unit 23 writes a plurality of detection points to the memory 12.

(Step S15: Approximation process)
The approximation unit 24 calculates an approximate straight line that approximates the movement locus obtained by connecting the plurality of detection points set in step S14 in time series.
Specifically, the approximation unit 24 reads a plurality of detection points from the memory 12. As shown in FIG. 6, the approximation unit 24 calculates an approximation straight line by the least squares method with the coordinates of a plurality of detection points as inputs. The method is not limited to the least squares method, and an approximate straight line that approximates the movement locus obtained by connecting a plurality of detection points in a time series may be calculated by another method. The approximation unit 24 writes the calculated approximate straight line to the memory 12.

(Step S16: Judgment process)
The determination unit 25 determines the average distance, which is the average value of the distances between the approximate straight line and each of the plurality of detection points, and the number of intersections of the movement locus obtained by connecting the approximate straight line and the plurality of detection points in a time series. Based on this, it is determined whether or not to notify. The determination process will be described in detail later.
When the determination unit 25 determines that the notification is to be made, the determination unit 25 writes the target person as a notification target in the memory 12.

(Step S17: Notification processing)
The notification unit 26 notifies the administrator or the like of the target person to be notified via the communication interface 14.
Specifically, the notification unit 26 reads the target person to be notified from the memory 12. The notification unit 26 transmits image data such as enclosing the target person to be notified in the latest captured image data with a frame to the display device, and displays the image data on the display device.

The determination process (step S16 in FIG. 2) according to the first embodiment will be described with reference to FIG. 7.
(Step S21: Distance determination process)
The determination unit 25 calculates the distance between the approximate straight line calculated in step S15 and each of the plurality of detection points set in step S14. As shown in FIG. 8, the distance between the approximate straight line and the detection point is the length of the perpendicular line drawn from the detection point to the approximate straight line. The determination unit 25 determines whether or not the average distance, which is the average value of the calculated distances, is equal to or greater than the first threshold value. The first threshold value is determined in advance depending on how large the meandering is to cause wobbling.
When the average distance is equal to or greater than the first threshold value, the determination unit 25 advances the process to step S22. On the other hand, when the average distance is less than the first threshold value, the determination unit 25 advances the process to step S24.

(Step S22: Crossing count determination process)
The determination unit 25 calculates the number of intersections between the approximate straight line calculated in step S15 and the movement locus obtained by connecting the plurality of detection points set in step S14 in time series. The determination unit 25 determines whether or not the number of intersections is equal to or greater than the second threshold value. The second threshold value is determined in advance according to how much meandering causes wobbling.
When the number of intersections is equal to or greater than the second threshold value, the determination unit 25 advances the process to step S23. On the other hand, when the number of intersections is less than the second threshold value, the determination unit 25 advances the process to step S24.

(Step S23: Target setting process)
The determination unit 25 determines that the target person is staggering, and writes the target person to the memory 12 as a notification target.

(Step S24: Non-target setting process)
The determination unit 25 determines that the target person is not staggering, and does not set the target person as the notification target.

*** Effect of Embodiment 1 ***
As described above, the notification determination device 10 according to the first embodiment sets the average distance, which is the average value of the distances between the approximate straight line and each of the plurality of detection points, and the number of intersections between the approximate straight line and the movement locus. Based on this, it is determined whether or not the target person is staggering.
Since it is determined whether or not there is wobbling by using not only the distance between the approximate straight line and the detection point but also the number of intersections between the approximate straight line and the movement locus, it is possible to appropriately detect the wobble. As a result, it is possible to appropriately notify a person who is staggering.

If only the distance between the approximate straight line and the detection point is used, it may be erroneously detected as staggering when moving in an arc shape while avoiding other people. However, by using the number of intersections, it is determined whether or not the vehicle meanders to the left or right while going from the start point to the end point. Therefore, it is possible to appropriately notify a person who is staggering.

*** Other configurations ***
<Modification example 1>
In the first embodiment, each functional component is realized by software. However, as a modification 1, each functional component may be realized by hardware. The difference between the first modification and the first embodiment will be described.

The configuration of the notification determination device 10 according to the first modification will be described with reference to FIG.
When each functional component is realized by hardware, the notification determination device 10 includes an electronic circuit 15 instead of the processor 11, the memory 12, and the 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.

Examples of the electronic circuit 15 include a single circuit, a composite circuit, a programmed processor, a parallel programmed processor, a logic IC, a GA (Gate Array), an ASIC (Application Specific Integrated Circuit), and an FPGA (Field-Programmable Gate Array). is assumed.
Each functional component may be realized by one electronic circuit 15, or each functional component may be realized by being distributed in a plurality of electronic circuits 15.

<Modification 2>
As a modification 2, some functional components may be realized by hardware, and other functional components may be realized by software.

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

Embodiment 2.
The second embodiment is different from the first embodiment in that it determines whether or not it is staggering in consideration of the crossing ratio, which is the ratio of the number of crossings to the moving distance. In the second embodiment, these different points will be described, and the same points will be omitted.

*** Explanation of operation ***
The operation of the notification determination device 10 according to the second embodiment will be described with reference to FIGS. 10 and 11.
The operation procedure of the notification determination device 10 according to the second embodiment corresponds to the notification determination method according to the second embodiment. Further, the program that realizes the operation of the notification determination device 10 according to the second embodiment corresponds to the notification determination program according to the second embodiment.

The determination process (step S16 in FIG. 2) according to the second embodiment will be described with reference to FIG.
The process from step S32 to step S33 is the same as the process from step S21 to step S22 in FIG. The process from step S35 to step S36 is the same as the process from step S23 to step S24 in FIG.

(Step S31: Movement distance determination process)
As shown in FIG. 11, the determination unit 25 calculates the moving radius, which is the radius of the smallest circle including the plurality of detection points set in step S14. That is, the determination unit 25 draws the smallest circle including all the plurality of detection points set in step S14, and calculates the distance between the center of the circle and the farthest detection point as the moving radius. Here, the moving radius is used as the moving distance specified from a plurality of detection points. The determination unit 25 determines whether or not the moving radius is equal to or greater than the shortest distance.
If the moving radius is equal to or greater than the shortest distance, the determination unit 25 advances the process to step S32. On the other hand, if the moving radius is less than the shortest distance, the determination unit 25 advances the process to step S36.

(Step S34: Crossing ratio determination process)
The determination unit 25 calculates the crossing ratio by dividing the number of crossings calculated in step S33 by the moving radius calculated in step S31. The determination unit 25 determines whether or not the intersection ratio is equal to or greater than the third threshold value. The third threshold value is determined in advance according to how much meandering per unit distance causes wobbling.
When the intersection ratio is equal to or higher than the third threshold value, the determination unit 25 advances the process to step S35. On the other hand, when the intersection ratio is less than the third threshold value, the determination unit 25 advances the process to step S36.

*** Effect of Embodiment 2 ***
As described above, the notification determination device 10 according to the second embodiment determines whether or not the device is staggering in consideration of the intersection ratio, which is the ratio of the number of intersections to the moving distance. Specifically, as shown in FIG. 12, when the number of crossings exceeds the second threshold value while traveling a short distance, it is detected as wobbling. On the other hand, if the number of crossings exceeds the second threshold value while traveling a long distance, it may not be detected as wobbling.
This makes it possible to detect wobbling more appropriately. For example, when moving while avoiding several people while running, there is a possibility of moving a long distance while moving left and right several times in a short time. In such a case, the notification determination device 10 according to the first embodiment may erroneously detect it as a wobble. On the other hand, the notification determination device 10 according to the second embodiment may be able to prevent erroneous detection as wobbling.

Whether or not an operation like this example should be detected as wobbling depends on the place where the photographing device 41 is installed and the like. Therefore, it may be possible to switch whether or not to use the function described in the second embodiment.

Embodiment 3.
In the third embodiment, a method of reducing the influence of the false detection in the detection process (step S12 of FIG. 2) will be described. In the third embodiment, only the points different from the first and second embodiments will be described, and the same points will be omitted.
Here, a case where the function is added to the first embodiment will be described. However, it is also possible to add a function to the second embodiment.

*** Explanation of operation ***
The operation of the notification determination device 10 according to the third embodiment will be described with reference to FIGS. 13 to 15.
The operation procedure of the notification determination device 10 according to the third embodiment corresponds to the notification determination method according to the third embodiment. Further, the program that realizes the operation of the notification determination device 10 according to the third embodiment corresponds to the notification determination program according to the third embodiment.

In the third embodiment, three methods 1 to 3 will be described as methods for reducing the influence of erroneous detection.

** Method 1 **
In the detection point setting process (step S14 in FIG. 2), the detection point setting unit 23 is a point where the distance from the set of points representing the position of the target person to the immediately preceding point in the time series is more than the limit distance. A plurality of points obtained by excluding the above are set as a plurality of detection points.
That is, as shown in FIG. 13, point 1, point 2, ... .. .. , It is assumed that a point indicating the coordinates of the target person is detected in the order of point 6. At this time, the point 4 is far from the immediately preceding point 3, and is separated by the limit distance or more. Such a point is likely to be a point that was erroneously detected in the detection process. Therefore, if the wobble is detected including such a point, the wobble may be erroneously detected. Therefore, such points are excluded from the set of points representing the position of the target person, and then a plurality of detection points are set. Therefore, in the case of the example shown in FIG. 13, points 1, point 2, point 3, point 5, and point 6 are a plurality of detection points. In this case, the process is performed assuming that the device has moved to the point 5 after the point 3.

** Method 2 **
In the determination process (step S16 in FIG. 2), the determination unit 25 moves in one direction in the depth direction of the captured image data, and then moves a distance of more than half of the distance moved in one direction in the opposite direction in the depth direction. If the movement locus indicates that, it is determined that the notification is not performed.
That is, as shown in FIG. 14, when a plurality of detection points are connected in chronological order, the movement locus may be such that the movement locus moves in the Y-axis direction (depth direction) and then returns to the original position. This may be the result of determining that two people are the same person when two people who move in one direction in the depth direction and a person who moves in the other direction in the depth direction pass each other. Therefore, in the case of such a movement locus, the determination unit 25 determines that there is no wobbling and determines that it does not notify.

** Method 3 **
In the crossing number determination process (step S22 in FIG. 7), the determination unit determines the detection point between the first intersection where the movement locus intersects the approximate straight line and the second intersection where the movement locus intersects the approximate straight line next. If the number of is less than the reference number for the number of detected points, the second intersection is not counted as the number of intersections.
That is, as shown in FIG. 15, it is assumed that a movement locus connecting a plurality of detection points is formed. At this time, the movement locus from the detection point 1 to the detection point 2 intersects the approximate straight line. Immediately after this, the movement locus from the detection point 2 to the detection point 3 intersects the approximate straight line. Immediately after that, the movement locus from the detection point 3 to the detection point 4 intersects the approximate straight line. In this way, once the movement locus intersects the approximate straight line, only the detection points smaller than the reference number for the number of detected points are sandwiched between them, and when the movement locus intersects the approximate straight line again, the point where another person is detected. May be included in multiple detection points. In the example of FIG. 15, as a result of determining that two people moving in parallel are the same person, there is a possibility that points indicating the positions of the two people are alternately connected. Therefore, in the case of such a movement locus, the crossing after crossing once is not counted in the number of crossings. Therefore, in the example shown in FIG. 15, only once the movement locus from the detection point 1 to the detection point 2 intersects the approximate straight line is counted as the number of intersections.

*** Effect of Embodiment 3 ***
As described above, the notification determination device 10 according to the third embodiment reduces the influence of erroneous detection in the detection process.
This makes it possible to detect wobbling more appropriately.

Embodiment 4.
The fourth embodiment is different from the first to third embodiments in that the wobbling is detected after correcting the distortion in the captured image data. In the fourth embodiment, these different points will be described, and the same points will be omitted.

*** Explanation of operation ***
The operation of the notification determination device 10 according to the fourth embodiment will be described with reference to FIGS. 16 to 18.
The operation procedure of the notification determination device 10 according to the fourth embodiment corresponds to the notification determination method according to the fourth embodiment. Further, the program that realizes the operation of the notification determination device 10 according to the fourth embodiment corresponds to the notification determination program according to the fourth embodiment.

The processing flow of the notification determination device 10 according to the fourth embodiment will be described with reference to FIG.
The processing of steps S41 to S43 is the same as the processing of steps S11 to S13 of FIG. The process from step S45 to step S48 is the same as the process from step S14 to step S17 in FIG.

(Step S44: Correction process)
The detection point setting unit 23 corrects the lens distortion of the photographing device 41 and converts the photographed image data into a bird's-eye view to generate the bird's-eye view image data.
Specifically, as shown in FIG. 17, the captured image data has a distortion that is stretched toward the four corners due to the influence of the wide-angle lens of the photographing device 41. Therefore, the detection point setting unit 23 corrects this distortion. The correction can be performed by setting parameters according to the photographing device 41 for the existing technique.
Further, as shown in FIG. 18, the photographing device 41 may photograph the photographing area from an oblique angle. Therefore, the detection point setting unit 23 converts the captured image data into a bird's-eye view and converts it into the bird's-eye view image data viewed from above. The bird's-eye view conversion can be performed by the existing technology.

In step S45, the detection point setting unit 23 sets a set of points representing the position of the target person in the bird's-eye view image data corrected in step S44 as a plurality of detection points.

*** Effect of Embodiment 4 ***
As described above, the notification determination device 10 according to the fourth embodiment corrects the distortion of the captured image data and then detects the wobbling. As a result, the positional relationship between the plurality of detection points becomes accurate and more appropriate. It is possible to detect wobbling.

The embodiments and modifications of the present invention have been described above. Some of these embodiments and modifications may be combined and carried out. In addition, any one or several may be partially carried out. The present invention is not limited to the above embodiments and modifications, and various modifications can be made as needed.

10 Notification judgment device, 11 Processor, 12 Memory, 13 Storage, 14 Communication interface, 15 Electronic circuit, 21 Image acquisition unit, 22 Detection unit, 23 Detection point setting unit, 24 Approximation unit, 25 Judgment unit, 26 Notification unit, 31 Position storage unit, 41 imaging device.

Claims (10)

An approximation part that calculates an approximate straight line that approximates the movement locus obtained by connecting multiple detection points representing the position of the subject detected in the reference period in time series, and
Whether to notify based on the average distance which is the average value of the distances between the approximate straight line and each of the plurality of detection points calculated by the approximate unit and the number of intersections between the approximate straight line and the movement locus. A notification determination device including a determination unit for determining whether or not. The notification determination device according to claim 1, wherein the determination unit determines to notify when the average distance is equal to or greater than the first threshold value and the number of intersections is equal to or greater than the second threshold value. The determination unit has the intersection at a moving radius that is the radius of the smallest circle including the plurality of detection points, the average distance of which is equal to or greater than the first threshold value and the number of intersections of which is equal to or greater than the second threshold value. The notification determination device according to claim 1, wherein the notification is determined when the crossing ratio divided by the number of times is equal to or greater than the third threshold value. The notification determination device further
A detection unit that detects the target person from the target photographed image data for each of the photographed image data of a plurality of frames obtained by photographing the photographing area with the photographing device.
Claims 1 to 3 include a detection point setting unit that sets a set of points representing the position of the target person detected by the detection unit from each of the captured image data of the plurality of frames as the plurality of detection points. The notification determination device according to any one of the above items. The detection point setting unit obtains a plurality of points obtained by excluding points that are separated from a set of points representing the position of the target person by a distance of a limit distance or more from the immediately preceding points in the time series. The notification determination device according to claim 4, which is set as the plurality of detection points. When the movement locus indicates that the determination unit has moved in one direction in the depth direction of the captured image data and then has moved a distance of half or more of the distance moved in the one direction in the opposite direction of the depth direction. The notification determination device according to claim 4 or 5, wherein it is determined not to notify. In the determination unit, the number of detection points between the first intersection where the movement locus intersects the approximate straight line and the second intersection where the movement locus intersects the approximate straight line is the reference number regarding the number of detection points. The notification determination device according to any one of claims 1 to 6, wherein the second intersection is not counted as the number of intersections when the number is less. The detection point setting unit corrects the lens distortion of the photographing device and generates bird's-eye view image data by performing bird's-eye view conversion for each of the photographed image data of the plurality of frames. The notification determination device according to claim 4 or 5, wherein a set of points representing the position of the target person in the bird's-eye view image data is set as the plurality of detection points. The approximation part calculates an approximate straight line that approximates the movement locus obtained by connecting a plurality of detection points representing the positions of the subjects detected in the reference period in chronological order.
The determination unit determines whether or not to notify based on the average distance which is the average value of the distances between the approximate straight line and each of the plurality of detection points and the number of intersections between the approximate straight line and the movement locus. Notification judgment method to be performed. An approximation part that calculates an approximate straight line that approximates the movement locus obtained by connecting multiple detection points representing the position of the subject detected in the reference period in time series, and
Whether to notify based on the average distance which is the average value of the distances between the approximate straight line and each of the plurality of detection points calculated by the approximate unit and the number of intersections between the approximate straight line and the movement locus. A notification determination program that causes a computer to function as a notification determination device that performs a determination unit that determines whether or not.

Images