Video monitoring device, program, and monitoring area setting method

Abstract

A video monitoring device (120) comprises: a marker detection unit (123) that extends in the length direction and that, by moving a continuous body provided with a marker, detects the marker from a video obtained by imaging the movement of the marker; and a setting unit (124) that sets a monitoring area in a portion of the video in which area monitoring is performed, so that a trajectory along which the marker moves is included. The setting unit (124) proportionately increases the monitoring area at a position where the marker is captured as the size of the marker in the video increases.

Description

Video monitoring device, program, and monitoring area setting method

[0001] The present disclosure relates to a video monitoring device, a program, and a monitoring area setting method.

[0002] In recent years, surveillance cameras have been installed in all sorts of places, including public places such as roads and construction sites, and automatic surveillance systems have been developed that use the images captured by the surveillance cameras and anomaly detection algorithms.

[0003] When applying an anomaly detection algorithm to video footage, it is common to reduce wasted computing resources and false positives by narrowing down the area to be monitored. In permanently installed surveillance cameras, the monitoring area is manually set by selecting points or areas on the image. In this case, once the monitoring area is set during initial setup, no further configuration is required, so the labor cost is low compared to the overall operation of the video surveillance system and is not a major issue.

[0004] However, algorithms that do not require long-term learning have been developed in recent years, and there has been an increase in video surveillance systems in which cameras are installed on-site and operated for only a short period of time after the work being monitored is completed. In such cases, the effort required to set up the surveillance area accounts for a large proportion of the overall operation, so it is necessary to be able to set up the surveillance area as automatically as possible.

[0005] As a technology for automatically setting a monitoring area, the system described in Patent Document 1 uses a monitoring area setting device that takes advantage of the fact that three markers are installed at equal intervals on the horizontal surface of the monitored object to calculate camera parameters and set the area surrounded by a straight line connecting the three markers and a straight line parallel to this at a regular interval as the monitoring area.

[0006] Japanese Patent Application Laid-Open No. 2001-145095

[0007] However, conventional technologies require the installation of multiple markers at equal intervals for each camera, limiting their scope of application. Furthermore, conventional technologies can only set a monitoring area on a horizontal plane and within a straight line. Therefore, conventional technologies cannot be applied when a curved surface in real space must be set as a monitoring area, such as when monitoring a factory line or cables during cable extension work. Furthermore, when targeting a continuous object such as a cable, it is difficult to confirm from the video that the object is moving in a continuous direction.

[0008] Therefore, according to one or more aspects of the present disclosure, it is an object of the present invention to reduce the effort required to install multiple markers and to enable the setting of a monitoring area on an image in a free shape that takes depth into consideration.

[0009] A video surveillance device according to one aspect of the present disclosure includes a marker detection unit that detects the marker from video footage of the movement of the marker by moving a continuum that extends in the longitudinal direction and is equipped with the marker; and a setting unit that sets a surveillance area, which is an area to be monitored in a portion of the video, so as to include the trajectory of the marker's movement, and the setting unit is characterized in that the larger the size of the marker in the video, the larger the surveillance area at the position where the marker is imaged.

[0010] A program according to one aspect of the present disclosure causes a computer to function as a marker detection unit that detects the movement of a marker from an image captured by moving a continuum that extends in a longitudinal direction and has a marker, and a setting unit that sets a monitoring area, which is an area to be monitored in a part of the image, so that the trajectory of the movement of the marker is included, and the setting unit is characterized in that the larger the size of the marker in the image, the larger the monitoring area at the position where the marker is captured.

[0011] A monitoring area setting method according to one aspect of the present disclosure comprises detecting a marker from an image capturing the movement of the marker by moving a continuum extending in a longitudinal direction and having the marker, and setting a monitoring area, which is an area to be monitored in a portion of the image, so that the trajectory of the marker's movement is included, and is characterized in that the larger the size of the marker in the image, the larger the monitoring area at the position where the marker is captured.

[0012] According to one or more aspects of the present disclosure, it is possible to reduce the effort required to install multiple markers and set a monitoring area on an image in a free shape that takes depth into consideration.

[0013] FIG. 1 is a block diagram schematically illustrating the configuration of a monitoring system according to a first embodiment. FIG. 2 is a block diagram schematically illustrating the configuration of a video monitoring device according to the first embodiment. (A) to (C) are schematic diagrams illustrating the operation of a marker detection unit. (A) to (C) are schematic diagrams illustrating the operation of a moving marker detection unit. (A) and (B) are schematic diagrams illustrating the operation of a marker trajectory identification unit. (A) and (B) are schematic diagrams illustrating the operation of a marker width calculation unit and a monitoring area setting unit. (A) and (B) are schematic diagrams illustrating an example of setting a monitoring area when a continuum passes through a curved location. FIG. 1 is a block diagram schematically illustrating the configuration of a PC. FIG. 2 is a block diagram schematically illustrating the configuration of a video monitoring device according to a second embodiment. (A) to (C) are schematic diagrams illustrating the operation of a marker width calculation unit. (A) and (B) are schematic diagrams illustrating the operation of a monitoring area setting unit. FIG. 3 is a schematic diagram illustrating an example of applying an anomaly detection algorithm to one frame included in a video.

[0014] 1 is a block diagram showing a schematic configuration of a monitoring system 100 according to embodiment 1. The monitoring system 100 includes a camera 110 and a video monitoring device 120. The camera 110 and the video monitoring device 120 are connected to a network 101 such as the Internet or a LAN (Local Area Network).

[0015] The monitoring system 100 here is a system for monitoring a cable laying method in which a continuous cable 102 is laid along a railway track 103. The cable 102 here is usually too heavy to be carried by hand and is laid over a relatively long distance. Specifically, the cable 102 can be laid by installing a wire laying machine 104 with two powered rollers 104a, 104b at appropriate intervals and feeding out the cable 102 from the front. The cable 102 is an example of a continuous body extending in the longitudinal direction.

[0016] The cable 102 is provided with a marker 105 for checking its movement. The marker 105 may be applied to any continuum, and is not limited to the cable 102. The marker 105 can also be moved by moving the cable 102. The marker 105 may be attached to the continuum, or may be printed on it.

[0017] The camera 110 is an imaging device that captures video for monitoring the status of work during the cable laying method. The captured video is provided to the video monitoring device 120 via the network 101. The camera 110 may be any type of camera, such as a color camera, grayscale camera, near-infrared camera, infrared camera, fisheye camera, or wide-angle camera, as long as it can capture video in which the marker 105 can be identified.

[0018] 2 is a block diagram showing a schematic configuration of the video monitoring device 120 according to the first embodiment. The video monitoring device 120 includes a communication unit 121, a video acquisition unit 122, a marker detection unit 123, and a setting unit 124. The setting unit 124 includes a moving marker detection unit 125, a marker trajectory identification unit 126, a marker width calculation unit 127, and a monitoring area setting unit 128.

[0019] The communication unit 121 performs communication via the network 101. For example, the communication unit 121 receives video from the camera 110 via the network 101. The received video is provided to the video acquisition unit 122.

[0020] The image acquisition unit 122 acquires an image from the camera 110. Here, the image acquisition unit 122 acquires the image from the camera 110 via the communication unit 121. The acquired image is provided to the marker detection unit 123. Note that if the image from the camera 110 has already been stored in a storage unit (not shown) or a storage device on the cloud, the image acquisition unit 122 may acquire the image from the storage unit or the storage device.

[0021] The marker detection unit 123 detects the marker from a video image capturing the movement of the marker by moving the cable 102. For example, the marker detection unit 123 extracts a marker area, which is an area in which the marker 105 appears, from each frame included in the video image, and generates a marker image showing the marker area.

[0022] 3A to 3C are schematic diagrams for explaining the operation of the marker detection unit 123. The marker detection unit 123 specifies the HSV color space or RGB color space range for each pixel of the frame 140A shown in Fig. 3A, and binarizes the frame 140A with a specific color (e.g., red) used as the color of the marker 105, thereby generating a binarized image 140B as shown in Fig. 3B.

[0023] Then, the marker detection unit 123 generates a marker image 140C as shown in Fig. 3C by leaving only regions of the binarized image 140B that have a certain area or more that allows the marker 105 to be distinguished from the noise 107. Note that the detection of the marker 105 may be performed by any known technique such as pattern matching as long as the outline of the marker can be recognized.

[0024] 2 , the setting unit 124 sets a monitoring area, which is an area to be monitored in a part of the video, so as to include the trajectory of the marker movement. Here, the setting unit 124 sets a larger monitoring area at the position where the marker is captured, as the size of the marker in the video increases. In the first embodiment, a first example of setting such a monitoring area is shown.

[0025] The moving marker detection unit 125 sequentially identifies two frames from multiple frames included in the video, and sequentially identifies moving marker areas indicating areas through which the two markers have moved, based on the difference between the two markers included in the two frames. The two frames may be sequentially identified so that they have the same time interval, for example.

[0026] For example, the moving marker detection unit 125 generates a moving marker image by extracting only the portion of the marker that has moved from the marker image.

[0027] 4A to 4C are schematic diagrams illustrating the operation of the moving marker detection unit 125. The moving marker detection unit 125 acquires a marker image 141A as shown in Fig. 4A. The marker image 141A is generated from a frame at a certain point in time, and the marker image 141A shows a marker area 141a, which is the area of ​​the marker.

[0028] As shown in Fig. 4(B), the moving marker detection unit 125 calculates the difference between the marker image 141A in Fig. 4(A) and a marker image 141B generated from a frame later than the marker image 141A in Fig. 4(A), thereby generating a moving marker image 142 indicating moving marker areas 142a and 142b, which are the moved portions of the marker area, as shown in Fig. 4(C). The generated moving marker image 142 is provided to the marker trajectory identification unit 126.

[0029] 2 , the marker trajectory identification unit 126 identifies a reference point within a moving marker area, and then identifies a reference line by connecting the multiple reference points identified according to the multiple frames. The marker trajectory identification unit 126 also identifies a trajectory area by combining the multiple moving marker areas identified according to the multiple frames.

[0030] For example, the marker trajectory identification unit 126 uses the moving marker image to create a marker trajectory image that indicates a trajectory region, which is the region of the trajectory along which the marker has moved.

[0031] 5A and 5B are schematic diagrams illustrating the operation of the marker trajectory identification unit 126. First, as shown in FIG. 5A, the marker trajectory identification unit 126 identifies reference points Pa and Pb within moving marker regions 142Aa and 142Ab indicated by the moving marker image 142A. Here, the reference points Pa and Pb are assumed to be the centers of gravity of the moving marker regions 142Aa and 142Ab. Then, as shown in FIG. 5B, a reference line RL is identified by connecting the reference points of multiple moving marker images generated from multiple frames included in the video. Note that when connecting the reference points to identify the reference line RL, either the reference point Pa or the reference point Pb may be used, or both the reference point Pa and the reference point Pb may be used to ensure stable operation.

[0032] In addition, the marker trajectory identification unit 126 identifies the trajectory region TR of the moving marker as shown in Figure 5 (B) by taking the logical sum of multiple moving marker images generated from multiple frames included in the video in the time axis direction.

[0033] The marker trajectory identification unit 126 generates a reference image 143 indicating the identified reference line RL and trajectory region TR and provides the reference image 143 to the marker width calculation unit 127 and the monitoring area setting unit 128. Instead of generating the reference image 143, the marker trajectory identification unit 126 may notify the marker width calculation unit 127 and the monitoring area setting unit 128 of the positions within the image of the identified reference line RL and trajectory region TR. In this case, the marker width calculation unit 127 and the monitoring area setting unit 128 may generate the reference image 143, or may draw only the trajectory region TR. Furthermore, in order to eliminate noise, the marker trajectory identification unit 126 may apply a noise removal process that leaves only regions having an area equal to or greater than a certain level. The marker trajectory identification unit 126 does not need to use all frames included in the video, and may skip frames within a range that does not affect the processing results.

[0034] 2 , the marker width calculation unit 127 calculates the distance from a sample point on the reference line to the boundary of the trajectory area in a direction intersecting the direction of marker movement as the marker width. Note that when the moving marker image moves from the foreground to the background in the image composition, the marker width gradually narrows.

[0035] For example, the marker width calculation unit 127 samples a plurality of points on the reference line identified by the marker trajectory identification unit 126, and calculates the marker width at each point.

[0036] 6 is a schematic diagram for explaining the operation of the marker width calculation unit 127 and the monitoring area setting unit 128. As shown in FIG. 6, the marker width calculation unit 127 calculates the distance l from the boundary of the trajectory region TR at the sample point p(x, y) in the direction perpendicular to the reference line RL. （ｘ、ｙ） is calculated as the marker width. （ｘ、ｙ） is notified to the monitoring area setting unit 128.

[0037] The marker width calculation unit 127 may calculate the width of the monitoring area separately for the left and right sides of the movement direction of the continuum. This makes it possible to set the monitoring area according to the object to be monitored, for example, when work on the continuum is performed only from one side. Furthermore, if the reference line is separated on the left and right sides of the boundary of the trajectory image, the monitoring area setting unit 128 can eliminate the separation and set the monitoring area by selecting marker widths separately for the left and right sides.

[0038] Returning to Fig. 2, the monitoring area setting unit 128 sets the monitoring area to include the trajectory region. Here, the monitoring area setting unit 128 sets the width of the monitoring area in the direction intersecting the reference line at the position of a sample point so that the width increases as the marker width at that sample point increases. Note that the monitoring area setting unit 128 sets the width of the monitoring area in the direction intersecting the reference line at the position of that sample point by multiplying the marker width at the position of that sample point by a predetermined value.

[0039] For example, the monitoring area setting unit 128 calculates a monitoring area that includes the reference line and in which the greater the marker width at a sample point, the greater the distance from the sample point to the boundary of the monitoring area.

[0040] As shown in FIG. 6, the monitoring area setting unit 128 sets the marker width l in a direction perpendicular to the reference line RL. （ｘ，ｙ） The area expanded by a constant n times (n>0) is set as the monitoring area AR.

[0041] As shown in FIG. 6, the width L of the monitoring area at point p(x, y) （ｘ，ｙ） is expressed by the following equation (1): As a result, the expansion width is adjusted by the arbitrary constant n.

[0042] L （ｘ，ｙ） = nl （ｘ，ｙ） (1)

[0043] Here, the marker width l is set on both sides of the reference line RL. （ｘ，ｙ） The example is explained using the case where the size of the marker width l is the same on both sides of the reference line RL, as described above. （ｘ，ｙ） The size of the monitoring area may be different, and the width L （ｘ，ｙ） For example, as expressed by the following formulas (2) and (3), the expansion width adjustment constant may be set to the left constant n Ｌ and the constant n on the right Ｒ By holding the left and right separately, the width of the monitoring area on each side is L Ｌ（ｘ，ｙ） and L Ｒ（ｘ，ｙ） can be adjusted.

[0044] L Ｌ（ｘ，ｙ） = n Ｌ l （ｘ，ｙ） (2) L Ｒ（ｘ，ｙ） = n Ｒ l （ｘ，ｙ） (3)

[0045] As described above, the monitoring area setting unit 128 can estimate the scale of the monitoring area on the image from the known actual marker width using the marker trajectory image generated by the marker trajectory identification unit 126 and the marker width calculated by the marker width calculation unit 127, and can set an area enlarged using the estimated scale as the monitoring area. （ｘ，ｙ） The larger the width of the monitoring area L （ｘ，ｙ） You can also enlarge the scale l （ｘ，ｙ） The smaller is, the （ｘ，ｙ） may be made smaller.

[0046] As described above, by taking advantage of the fact that a moving marker appears larger when it is closer and smaller as it moves farther away, it is possible to set a monitoring area of ​​any shape that takes depth into consideration from the position of the marker obtained by the marker trajectory identification unit 126 and the marker width calculation unit 127 and the change in size.

[0047] The width of the expansion, in other words, the constant n, may be different on the left and right sides of the continuum in the moving direction. Also, the trajectory of the marker does not necessarily have to be a straight line. As shown in Figure 7, the continuum may pass through a curved area.

[0048] By configuring the monitoring system 100 in this way, the time and effort required to install multiple markers can be reduced, and a monitoring area can be set on an image in any shape, taking depth into consideration.

[0049] The video monitoring device 120 described above can be realized by, for example, a computer such as the PC 10 shown in Fig. 8. The PC 10 includes a storage 11 such as a hard disk drive (HDD) and a solid state drive (SSD), a memory 12, a processor 13 such as a central processing unit (CPU), and a communication interface (I / F) 14 such as a network interface card (NIC). The PC 10 may also include an input I / F 15 such as a keyboard and a mouse, and a display 16.

[0050] For example, the video acquisition unit 122, the marker detection unit 123, the moving marker detection unit 125, and the setting unit 124 can be realized by the processor 13 executing a program. The communication unit 121 can be realized by the communication I / F 14.

[0051] The program may be downloaded to the storage 11 from a recording medium (not shown) via a reader / writer (not shown) or from the network 101 via the communication I / F 14, and then loaded onto the memory 12 and executed by the processor 13. Alternatively, the program may be directly loaded onto the memory 12 from a recording medium via the reader / writer or from the network 101 via the communication I / F 14, and then executed by the processor 13. In other words, the program may be provided by a computer program product such as a recording medium.

[0052] Second Embodiment As shown in FIG. 1, a monitoring system 200 according to a second embodiment includes a camera 110 and a video monitoring device 220 .

[0053] 9 is a block diagram showing a schematic configuration of a video monitoring device 220 according to embodiment 2. The video monitoring device 220 includes a communication unit 121, a video acquisition unit 122, a marker detection unit 123, and a setting unit 224. The setting unit 224 includes a moving marker detection unit 125, a monitoring area setting unit 228, and an area size calculation unit 229.

[0054] The communication unit 121, the video acquisition unit 122, the marker detection unit 123, and the moving marker detection unit 125 of the video monitoring device 220 in embodiment 2 are the same as the communication unit 121, the video acquisition unit 122, the marker detection unit 123, and the moving marker detection unit 125 of the video monitoring device 220 in embodiment 2. However, in embodiment 2, the moving marker detection unit 125 provides the moving marker image to the area size calculation unit 229 and the monitoring area setting unit 228.

[0055] The setting unit 224 sets a monitoring area, which is an area to be monitored in a part of the video, so that the trajectory of the marker movement is included. Here, the setting unit 224 sets a larger monitoring area at the position where the marker is captured, as the size of the marker in the video increases. In the second embodiment, a second example of setting such a monitoring area is shown.

[0056] The region size calculation unit 229 identifies a reference point included in the moving marker region shown in the moving marker image. In this case, the reference point is assumed to be the center of gravity of the moving marker region. The region size calculation unit 229 then calculates the distance from the reference point to the farthest point within the moving marker region as the region size. The calculated region size is notified to the monitoring area setting unit 228.

[0057] It is considered that the moving marker area increases as the size of the marker in the video increases, and therefore the larger the area size, the larger the marker size will be.

[0058] 10A to 10C are schematic diagrams for explaining the operation of the region size calculation unit 229. When a moving marker image 244 as shown in Fig. 10A is acquired, the region size calculation unit 229 specifies the center of gravity Pc as a reference point, and calculates the distance l between the center of gravity Pc and the farthest point, as shown in Fig. 10B. １ is specified as the region size.

[0059] The shape of the marker does not have to be rectangular. For example, even if the marker is a moving marker area as shown in FIG. 10C, the area size calculation unit 229 specifies the center of gravity Pd as a reference point, and calculates the distance l between the center of gravity Pd and the farthest point. ２ is specified as the region size.

[0060] 9 , the monitoring area setting unit 228 identifies multiple expanded moving marker areas from multiple frames by expanding the moving marker area so that the larger the area size, which is the size of the moving marker area, the larger the expanded moving marker area becomes.The monitoring area setting unit 228 then sets the monitoring area by combining the multiple expanded moving marker areas.

[0061] For example, the monitoring area setting unit 228 identifies the expanded moving marker area by expanding the moving marker area indicated by the moving marker image, and identifies the expanded marker trajectory area by taking the logical OR of the expanded moving marker area in the time axis direction corresponding to multiple frames included in the video.

[0062] A commonly used morphological process is applied as the dilation process. Specifically, the monitoring area setting unit 228 slides a specific kernel over the image and sets the new value of each pixel to the maximum value of its neighborhood. The kernel generally used at this time is a rectangular kernel, an elliptical kernel, a cross-shaped kernel, or the like, but any of them can be used. Here, the iteration, which is a parameter for specifying how many times the dilation process is applied, is a constant that is set by multiplying the area size l by the area size l. １ and l ２ It is determined by multiplying by

[0063] Then, the monitoring area setting unit 228 sets the expansion marker trajectory region as the monitoring area.

[0064] 11A and 11B are schematic diagrams illustrating the operation of the monitoring area setting unit 228. First, as shown in FIG. 11A, the monitoring area setting unit 228 specifies expanded moving marker areas 244c and 244d by expanding moving marker areas 244a and 244b indicated by a moving marker image 244. Then, as shown in FIG. 11B, the monitoring area setting unit 228 performs expansion processing on multiple moving marker images generated from multiple frames included in the video, and performs a logical OR on the expanded moving marker images to specify an expanded marker movement area 245a, which is the area to which the expanded marker has moved, and sets the expanded marker movement area 245a as the monitoring area. Note that the monitoring area setting unit 228 may skip frames to the extent that it does not affect the processing results.

[0065] As described above, the monitoring system 200 according to the second embodiment also reduces the effort required to install multiple markers and allows for the setting of a monitoring area of ​​any shape, taking depth into consideration.

[0066] Embodiment 3 The video monitoring device 120, 220 in the above-described embodiment 1 or 2 may be provided with a monitoring unit (not shown) that applies a known anomaly detection algorithm to partial video, which is a part of the video acquired by the video acquisition unit 122 and is within the monitoring area set by the monitoring area setting unit 128, 228, in order to detect anomalies.

[0067] By providing a monitoring unit as described above, it is possible to determine whether the worker is normal or abnormal from the video, thereby eliminating disturbance factors from the perspective of the abnormality detection algorithm, such as the passage of a worker.

[0068] 12 is a schematic diagram illustrating an example of applying an anomaly detection algorithm to one frame included in a captured video. As shown in FIG. 12, the video monitoring device 120, 220 can automatically set a monitoring area AR1 and non-monitoring areas AR2, AR3, thereby limiting monitoring by the monitoring unit to only the monitoring area AR1. This reduces the monitoring load on the monitoring unit.

[0069] As described above, in conventional technology, when setting up a monitoring area, it was necessary to place multiple markers at equal intervals on a straight line in the monitoring area. However, according to embodiments 1 to 3, by providing one marker on a moving continuum, it is possible to reduce the effort required to place multiple markers at a monitoring location.

[0070] Furthermore, in conventional technology, in order to set a monitoring area that takes into account the depth of the actual size, camera parameters are calculated by taking advantage of the fact that markers are placed at equal intervals on a straight line, and only areas on a horizontal plane separated by straight lines including the markers can be set as the monitoring area. In contrast, according to embodiments 1 to 3, the trajectories of the markers provided on a continuum and the size changes of the markers in the video can be used to set a monitoring area of ​​any shape, including a three-dimensional area.

[0071] Furthermore, while conventional technology required the installation of a marker for each surveillance camera, according to embodiments 1 to 3, the surveillance area can be automatically set by moving a continuum equipped with a marker, eliminating the need to install a marker for each camera.

[0072] 100, 200 Surveillance system, 110 Camera, 120, 220 Video surveillance device, 121 Communication unit, 122 Video acquisition unit, 123 Marker detection unit, 124, 224 Setting unit, 125 Moving marker detection unit, 126 Marker trajectory identification unit, 127 Marker width calculation unit, 128, 228 Surveillance area setting unit, 229 Area size calculation unit.

Claims

1. A video surveillance device comprising: a marker detection unit that detects the marker from an image of the movement of the marker by moving a continuum that extends in the length direction and has the marker; and a setting unit that sets a monitoring area, which is an area to be monitored in part of the image, so that the trajectory of the movement of the marker is included, wherein the setting unit increases the monitoring area at the position where the marker is captured as the size of the marker in the image increases.

2. The video monitoring device according to claim 1, wherein the setting unit comprises: a moving marker detection unit that sequentially identifies two frames from a plurality of frames included in the video, and sequentially identifies moving marker areas that indicate areas through which the two markers have moved based on the difference between the two markers included in the two frames; a marker trajectory identification unit that identifies a reference point within the moving marker area, thereby connecting the plurality of reference points identified according to the plurality of frames to identify a reference line, and combines the plurality of moving marker areas identified according to the plurality of frames to identify a trajectory area; a marker width calculation unit that calculates the distance from a sample point on the reference line to the boundary of the trajectory area in a direction intersecting the direction in which the marker moves, as a marker width; and a monitoring area setting unit that sets the monitoring area to include the trajectory area, and sets the width of the monitoring area in the direction intersecting the reference line at the position of the sample point so that the width of the monitoring area in the direction intersecting the reference line at the position of the sample point increases as the marker width at the sample point increases.

3. The video monitoring device according to claim 2, wherein the monitoring area setting unit sets the width of the monitoring area in the direction intersecting the reference line at the position of the sample point by multiplying the marker width at the position of the sample point by a predetermined value.

4. The video surveillance device of claim 1, wherein the setting unit comprises: a moving marker detection unit that sequentially identifies two frames from the multiple frames included in the video, and sequentially identifies moving marker areas indicating areas through which the two markers have moved based on the difference between the two markers included in the two frames; and a monitoring area setting unit that identifies multiple expanded moving marker areas from the multiple frames by expanding the moving marker areas so that the larger the area size that is the size of the moving marker area, the larger the expanded moving marker areas, and sets the monitoring area by combining the multiple expanded moving marker areas.

5. The video monitoring device according to claim 4, wherein the monitoring area setting unit specifies the distance from the center of gravity of the moving marker area to the farthest point within the boundary of the moving marker area as the area size.

6. A video monitoring device according to any one of claims 1 to 5, further comprising a monitoring unit for detecting abnormalities in the monitoring area.

7. A program that causes a computer to function as: a marker detection unit that detects the marker from an image of the movement of the marker by moving a continuum that extends in the length direction and has a marker; and a setting unit that sets a monitoring area, which is an area to be monitored in a part of the image, so that the trajectory of the movement of the marker is included, and the setting unit increases the monitoring area at the position where the marker is imaged as the size of the marker in the image increases.

8. A monitoring area setting method for detecting a marker from an image capturing the movement of the marker by moving a continuum that extends in the length direction and has a marker attached thereto, and setting a monitoring area, which is an area to be monitored in a part of the image, so that the trajectory of the marker's movement is included, characterized in that the larger the size of the marker in the image, the larger the monitoring area at the position where the marker is captured.

Images