JPWO2018207476A1 - Image processing apparatus, monitoring camera, monitoring system, program, and image processing method - Google Patents

Abstract

The monitoring accuracy is improved while suppressing and reducing the calculation cost with a simple configuration. A monitoring target detection unit (12) that detects whether or not the monitoring target (900) exists in the attention monitoring area, and a monitoring image processing execution unit (13) that performs monitoring image processing on the image (I) in the attention monitoring area. When the monitoring target detection unit (12) detects the presence of the monitoring target (900), the monitoring image processing execution unit (13) executes the monitoring image processing.

Description

  The present invention relates to an image processing device, a monitoring camera, a monitoring system, a program, and an image processing method.

  In recent years, research and development of surveillance cameras, surveillance systems, and the like for the purpose of crime prevention and security maintenance, such as detection of suspicious persons, have been actively conducted to improve the monitoring accuracy while suppressing the calculation cost. For example, Patent Document 1 discloses a surveillance system and the like having a surveillance camera and a flow line analyzer. The flow line analyzer is a flow line measurement unit that measures a flow line to be monitored, a visible region measurement unit that measures a visible region, and an overlap detection that detects a temporal and spatial overlap between the flow line and the visible region. Means, and a flow line extracting means for extracting a flow line of a section having a temporal / spatial overlap.

Japanese Unexamined Patent Publication "JP-A-2006-103690 (published on June 2, 2016)"

  However, the surveillance system disclosed in Patent Literature 1 requires not only a surveillance camera to be installed but also a surveillant to wear a wearable camera, and thus requires at least two cameras. Therefore, the configuration of the surveillance system is more complicated and more expensive than monitoring with a single surveillance camera. In addition, the traffic line analysis device disclosed in Patent Literature 1 is for a suspicious person or the like existing outside the visible region of the observer wearing the wearable camera, that is, outside the photographable range of the wearable camera. The flow line analysis cannot be performed. Therefore, especially when a suspicious person or the like exists outside the visible region of the monitor, the monitoring accuracy is reduced.

  An aspect of the present invention has been made in view of the above-described problems, and an object of the present invention is to improve monitoring accuracy while suppressing calculation costs and reducing costs with a simple configuration.

  In order to solve the above problem, an image processing device according to an aspect of the present invention is an image processing device that processes a recorded image recorded by a surveillance camera, and is set in a first area in the recorded image. A monitoring target detecting unit that detects whether or not a monitoring target exists in the target monitoring region; and executing a monitoring image process for performing monitoring image processing for monitoring the monitoring target with respect to an image in the target monitoring region. And a monitoring image processing execution unit that executes the monitoring image processing when the monitoring target detection unit detects the presence of the monitoring target.

  According to another embodiment of the present invention, there is provided an image processing method for processing a recorded image recorded by a surveillance camera, wherein the image processing method includes: A monitoring target detecting step of detecting whether or not a monitoring target exists in the set target monitoring region; and a monitoring image for performing monitoring image processing for monitoring the monitoring target with respect to an image in the target monitoring region. And performing a monitoring image process in the monitoring image processing execution step when the presence of the monitoring target is detected in the monitoring target detection step.

  According to one aspect of the present invention, the area within a recorded image on which monitoring image processing is performed is limited to a target monitoring area, so that monitoring accuracy is improved while suppressing and reducing calculation costs with a simple configuration. be able to.

FIG. 1 is a block diagram illustrating an example of a functional configuration of a monitoring system according to Embodiments 1, 2, 3, and 4 of the present invention. FIG. 2 is a block diagram illustrating an example of a functional configuration of a surveillance camera according to Embodiments 1, 2, 3, and 4 of the present invention. (A) is a schematic diagram showing an example of setting an attention monitoring area in the daytime. (B) is a schematic diagram showing an example of setting a watched monitoring area at night. It is the schematic which shows the relationship between the operation | movement of a suspicious person and the said attention monitoring area. 4 is a flowchart illustrating an example of an image processing method according to the first embodiment of the present invention. It is a block diagram showing the functional composition of the modification of the surveillance camera. It is the schematic which shows the example of a setting of the attention monitoring area in the daytime. 9 is a flowchart illustrating an example of an image processing method according to a second embodiment of the present invention. 13 is a flowchart illustrating an example of an image processing method according to a third embodiment of the present invention. 13 is a flowchart illustrating an example of an image processing method according to a fourth embodiment of the present invention.

[Embodiment 1]
Hereinafter, embodiments of the present invention will be described in detail with reference to FIGS. In the following embodiments, a case where an image processing apparatus according to one embodiment of the present invention is incorporated in a monitoring camera will be described as an example.

  Note that the image processing device according to one embodiment of the present invention does not necessarily need to be built in a surveillance camera. For example, the image processing apparatus according to one embodiment of the present invention is incorporated in an external information processing apparatus, and information related to a recorded image recorded by the monitoring camera is transmitted and received between the information processing apparatus and the monitoring camera, so that monitoring is performed. Image processing may be performed.

<Functional configuration of surveillance system and surveillance camera>
First, a functional configuration of the monitoring system 100 and the monitoring camera 1 will be described with reference to FIGS. FIG. 1 is a block diagram illustrating an example of a functional configuration of the monitoring system 100. FIG. 2 is a block diagram illustrating an example of a functional configuration of the monitoring camera 1. FIG. 3A is a schematic diagram showing an example of setting a monitoring area of interest in the daytime. FIG. 3B is a schematic diagram illustrating an example of setting a target monitoring area at night. FIG. 4 is a schematic diagram showing the relationship between the operation of the suspicious individual 900 and the attention monitoring area.

  The surveillance system 100 records the situation in the surveillance area 800 (see FIG. 3 and the like) by the surveillance camera 1 for a certain period of time or for a certain period of time. 900 (monitoring target: see FIG. 4). In the following embodiments (other than the fourth embodiment), it is assumed that the floor of a store where products are sold is set as the monitoring area 800.

  As the monitoring area 800 in which the monitoring system 100 can be introduced, in addition to the store floor as described above, spaces in various buildings and facilities are targeted. In each of the following embodiments, the monitoring target of the monitoring system 100 is the suspicious individual 900, but for example, a specific animal may be the monitoring target. In this case, the monitoring area 800 is assumed to be, for example, a store in a mountainous area that sells food.

  As shown in FIG. 1, the monitoring system 100 includes a monitoring camera 1, a wireless LAN 101, a monitoring server 102, and an external storage device 103. The monitoring camera 1 transmits information on each of the recorded image R and the monitored area of interest image I (see FIG. 4: details will be described later) to the monitoring server 102 and the external storage device 103 via the wireless LAN 101. Further, the monitoring server 102 performs various processes requested by a monitor (not shown) by using information relating to each of the recorded image R and the target monitoring area image I. For example, when it is determined that the suspicious individual 900 has been detected from the information on the attention monitoring area image I, the monitoring server 102 transmits information on the suspicious individual detection to a security company or the like. The external storage device 103 stores information on each of the recorded image R and the watched monitoring area image I. Further, in response to the call from the monitoring camera 1 or the like, the stored information is transmitted to the monitoring camera 1 or the like. As the external storage device 103, for example, a NAS (Network Attached Storage; network-attached storage), an HDR (HardDisk Recorder; hard disk recorder), or the like is assumed.

  As shown in FIG. 2, the monitoring camera 1 includes a photographing unit 5, a radio unit 6, a first power supply unit 7a, a second power supply unit 7b, a first memory 8a, a second memory 8b, and a control unit 9. The control unit 9 includes a first image processing device 10a (image processing device) and a second image processing device 10b (image processing device). The imaging unit 5 includes a lens, an image sensor, and the like, and records a state in the monitoring area 800 to generate a recorded image R. The wireless unit 6 is a wireless LAN circuit, and various types of information are transmitted and received between the monitoring camera 1 and the monitoring server 102 / external storage device 103 via the wireless unit 6.

  The first power supply unit 7a supplies power to the first image processing device 10a, and the second power supply unit 7b supplies power to the second image processing device 10b. The first memory 8a is an operation memory for the first image processing device 10a to perform various processes, and the second memory 8b is an operation memory for the second image processing device 10b to perform various processes. Further, both the first and second memories 8a and 8b store various information handled by the monitoring camera 1.

  The control unit 9 controls each unit of the monitoring camera 1. In FIG. 2, the control unit 9 is built in the surveillance camera 1, but the control unit 9 is an external device (not shown) external to the surveillance camera 1 or a network server used via the wireless unit 6. There may be.

  Both the first and second image processing apparatuses 10a and 10b are configured by, for example, CPUs. The first image processing device 10a performs a recording process of the entire monitoring area 800 and an image process of a recorded image R recorded by the photographing unit 5 of the monitoring camera 1, and includes a setting switching unit 11 and a monitoring target detecting unit 12. I have. The second image processing device 10b is a device that mainly performs monitoring image processing (details will be described later), and includes a monitoring image processing execution unit 13.

  The recording processing of the entire monitoring area 800 and the image processing of the recorded image R by the first image processing device 10a are each performed at the normal resolution. However, when the monitoring camera 1 operates in the low power consumption mode, each of these processes is performed at a low resolution.

  Here, in the present embodiment, as shown in FIG. 3A, when monitoring the monitoring area 800 in the daytime, the target monitoring area is initially set in the first area F11 and the first area F12 in the recorded image R. Is done. The first area F11 is a recorded image portion including a location of a cash register in the recorded image R, and the first area F12 is a recorded image portion including a display portion of a high-priced product in the recorded image R.

  The attention monitoring area is a recorded image portion in the recorded image R, which can be arbitrarily initialized by a monitor as a processing target of the monitoring image processing (details will be described later) by the monitoring image processing execution unit 13. Specifically, the target monitoring area may be initially set to an arbitrary portion in the recorded image R, and the size of the target monitoring area may be changed as appropriate. This initial setting may be performed in advance before the operation of the monitoring camera 1, or may be performed during the operation of the monitoring camera 1.

  The store, which is the monitoring area 800, is open during the daytime and many people enter and exit the store and the entrances and exits for employees. Therefore, specific persons at these locations behave suspiciously. Even so, it is difficult to discover. Therefore, it is difficult to detect the suspicious person 900 even if these parts are monitored in the daytime.

  On the other hand, in the vicinity of cash registers where cash is present, or near places where high-priced products are displayed where the damage to stores becomes more serious, even during the daytime, high-priced goods may be taken away or cash or high-priced goods may be robbed. There is. In addition, in the event of removal or the like, it is necessary to ensure the safety of employees at and around the site. Therefore, when monitoring the monitoring area 800 in the daytime, the target monitoring area is initially set in each of the first areas F11 and F12 in the recorded image R.

  Note that the above-described initial setting of the target monitoring area is merely an example, and the present invention is not limited to this case. For example, the target monitoring region may be initially set in two recorded image portions other than the first regions F11 and F12 in the recorded image R. Further, for example, the attention monitoring area may be initially set in one recorded image portion in the recorded image R. Furthermore, the number and the mode of the initially set target monitoring areas may be changed according to the characteristics of the location that becomes the monitoring area 800.

  The setting switching unit 11 illustrated in FIG. 2 determines whether or not each of the two switching conditions (the first condition) is satisfied. If it is determined that one of the switching conditions is satisfied, the setting switching unit 11 changes the target monitoring area in the recorded image R from the first area F11 / F12 to the second area F21 / F22 different from the first area F11 / F12. (See FIG. 3B). On the other hand, when determining that the other of the switching conditions is satisfied, the setting switching unit 11 returns the target monitoring area from the second areas F21 and F22 to the first areas F11 and F12 (see FIG. 3A).

  When the switching condition is satisfied, it is necessary to switch the monitoring area of interest in the recorded image R. Although various conditions are assumed as the switching conditions, in the present embodiment, a first specific time at which switching is performed from daytime to nighttime is one of the switching conditions. In this embodiment, 18:30 is set as the first specific time. Further, a second specific time at which switching is performed from night to day is the other of the switching conditions. In the present embodiment, 9:30 is set as the second specific time. The setting of these specific times may be input by an observer from an operation input unit (not shown) of the monitoring camera 1, or may be stored in the first memory 8a in advance.

  Also, as shown in FIG. 3B, the second area F21 is a recorded image portion including the entrance for the employee in the recorded image R, and the second area F22 is a store in the recorded image R. Is a recorded image portion including the entrance of the camera. For example, the information relating to the second areas F21 and F22 is stored in advance in the first memory 8a, and when it is determined that the setting switching condition is satisfied, the setting switching unit 11 calls up the information from the first memory 8a. Then, the target monitoring areas set in the first areas F11 and F12 are switched to the second areas F21 and F22.

  The reason why the switching condition is set to such specific times (9:30, 18:30) is that the store, which is the monitoring area 800, differs in the number of people entering and leaving and the characteristics of people entering and leaving according to the time zone. It is. In other words, many people including general customers enter and leave the store during the day, but the number of people entering and exiting the night decreases, so that it is possible to observe the situation of each person entering and exiting the store in more detail. In addition, since the ratio of employees to all persons who enter and exit the store at night increases, suspicious persons 900 are relatively mingled with those who enter the store at night except for employees as compared to daytime. The likelihood increases.

  Considering the characteristics of the daytime / nighttime monitoring area 800, in the nighttime, the focused monitoring area includes the recorded images near the entrance of the store or the entrance / exit for the employees, so that these areas are emphasized. If monitoring is performed, there is a high possibility that the suspicious individual 900 can be detected earlier. Therefore, one of the switching conditions is set to 18:30, which can be said to be the time zone at night. At 18:30, the attention monitoring area is switched to the second areas F21 and F22 for night, and the suspicious person 900 is set. Can be detected earlier. Note that the above-described switching of the attention monitoring area to the second areas F21 and F22 is merely an example, and is not limited to this case.

  The monitoring target detection unit 12 illustrated in FIG. 2 is configured to determine whether the suspicious individual 900 exists in the attention monitoring area initially set in the first areas F11 and F12 before the attention switching area is switched by the setting switching unit 11. Is detected (see FIG. 3A). Further, after the target monitoring region is switched by the setting switching unit 11, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the target monitoring region switched to the second region F21 or F22. (See FIG. 3B).

  Here, the case where the monitoring target detection unit 12 detects the presence of the suspicious individual 900 means that the suspicious individual 900 (specifically, at least a part of the body of the suspicious individual 900) is monitored as shown in FIG. This refers to a case where the monitoring target detection unit 12 recognizes that the image is displayed in the area (the first area F11 or F12 or the second area F21 or F22) (see the arrow on the paper surface in the drawing). On the other hand, the case where the monitoring target detection unit 12 does not detect the presence of the suspicious individual 900 indicates the case where the monitoring target detection unit 12 recognizes that the suspicious individual 900 is no longer displayed in the attention monitoring area (in the figure). See the arrow below the paper).

  The detection operation of the target monitoring area switched to the second areas F21 and F22 by the monitoring target detection unit 12 is performed by the setting switching unit 11 by changing the target monitoring area from the first areas F11 and F12 to the second areas F21 and F22. Is started by receiving information indicating that the switching has been performed.

  The monitoring image processing execution unit 13 executes monitoring image processing for monitoring the suspicious individual 900 on the attention monitoring area image I when the monitoring target detection unit 12 detects the presence of the suspicious individual 900 in the attention monitoring area. I do. Here, the execution of the monitoring image processing by the monitoring image processing execution unit 13 is started by receiving information indicating that the presence of the suspicious individual 900 has been detected from the monitoring target detection unit 12.

  Here, the attention monitoring area image I refers to a recorded image of an area in the attention monitoring area where the presence of the suspicious individual 900 is detected (see FIG. 4). In addition, as an example of the monitoring image processing, a process of cutting out only an area in which the presence of the suspicious individual 900 is detected in the attention monitoring area, recording the image at high resolution, and generating a high-resolution attention monitoring area image I will be described. be able to. Further, for example, a process of performing image analysis of the attention monitoring region image I and recognizing a face projected in the image I can be mentioned.

  Further, as the monitoring image processing, for example, the monitoring image processing execution unit 13 calls the suspicious individual database stored in the monitoring server 102 or an external information processing device or the like, and the image of the suspicious individual database and the target monitoring area image I And a process of collating with the face shown in FIG. Then, when the image of the suspicious individual database matches the face projected in the attention monitoring area image I, a process of determining the person projected in the image I as the suspicious individual 900 is also given as an example. Can be.

  Furthermore, until the monitoring target detection unit 12 detects the presence of the suspicious individual 900, the monitoring target area is recorded at a low resolution, and after the monitoring target detection unit 12 detects the presence of the suspicious individual 900 in the attention monitoring region, For example, a process of increasing the resolution of an area where the presence of the suspicious individual 900 has been detected and recording the image to generate the attention monitoring area image I can also be mentioned.

  In addition, the monitoring image processing execution unit 13 continues or stops the execution of the monitoring image processing according to the detection result regarding the presence or absence of the suspicious individual 900 by the monitoring target detection unit 12. Specifically, the monitoring image processing execution unit 13 starts execution of the monitoring image processing when the monitoring target detection unit 12 detects the presence of the suspicious person 900 in the attention monitoring area, and detects the presence of the suspicious person 900. During this time, the execution of the monitoring image processing is continued. On the other hand, when the monitoring target detection unit 12 stops detecting the presence of the suspicious individual 900, the monitoring image processing execution unit 13 stops executing the monitoring image processing.

<Image processing method by first image processing device and second image processing device>
Next, an image processing method by the first image processing device 10a and the second image processing device 10b will be described with reference to FIG. FIG. 5 is a flowchart illustrating an example of the image processing method.

  First, when the monitoring camera 1 starts the recording process of the entire monitoring area 800, the process of the flowchart shown in FIG. 5 is started (START). In step S101 (hereinafter, “step” is omitted), the monitoring image processing execution unit 13 starts the recording processing of the target monitoring area initially set in the first areas F11 and F12 in the recorded image R, and proceeds to step S102. Proceed to.

  Next, the setting switching unit 11 checks the time in S102, and determines in S103 whether the checked time satisfies one of the switching conditions (18:30). If NO (hereinafter, abbreviated as “N”) is determined in S103, the monitoring target detection unit 12 determines whether the suspicious individual 900 exists in the attention monitoring area initially set in the first areas F11 and F12. Is detected (S105: monitoring target detection step).

  On the other hand, if it is determined YES in S103 (hereinafter abbreviated as “Y”), the setting switching unit 11 changes the target monitoring area in the recorded image R from the first area F11 / F12 to the second area F21 in S104.・ Switch to F22 and proceed to S105. In S105, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring area switched to the second areas F21 and F22.

  If Y is determined in S105, the monitoring image processing execution unit 13 starts execution of the monitoring image processing for the attention monitoring area image I in the attention monitoring area in which the presence of the suspicious individual 900 is detected, and proceeds to S107. Proceed (S106: monitoring image processing execution step). On the other hand, if it is determined to be N in S105, the monitoring image processing execution unit 13 continues the recording processing of the target monitoring area switched to the second areas F21 and F22. In addition, the setting switching unit 11 executes the process of S102 again.

  Next, in S107, the monitoring target detection unit 12 detects again whether or not the suspicious individual 900 exists in the attention monitoring region (a monitoring target detection step). When Y is determined in S107, the execution of the monitoring image processing by the monitoring image processing execution unit 13 continues, and the monitoring target detection unit 12 executes the processing of S107 again.

  On the other hand, when it is determined to be N in S107, the monitoring image processing execution unit 13 stops executing the monitoring image processing in S108, and returns to S102. The setting switching unit 11 confirms the time in S102, and determines whether the confirmed time satisfies the other of the switching conditions (in 9:30) or not in S103. If it is determined to be N in S103, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring area that has been switched to the second areas F21 and F22 in S105.

  On the other hand, if it is determined to be Y in S103, the setting switching unit 11 returns the attention monitoring area in the recorded image R from the second areas F21 and F22 to the first areas F11 and F12 in S104, and proceeds to S105. Next, in S105, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the target monitoring area set again in the first areas F11 and F12.

  The processes in S106 to S108 are the same as described above, and the processes in S102 to S108 are continuously executed until the monitoring of the monitoring area 800 by the monitoring system 100 ends.

<Modification>
As the monitoring camera, the image processing device, and the monitoring system according to one embodiment of the present invention, not only the monitoring camera 1, the first and second image processing devices 10a and 10b, and the monitoring system 100 described above, but also a plurality of modified examples are assumed. This will be described with reference to FIG. FIG. 6 is a block diagram illustrating a functional configuration of a modified example of the monitoring camera 1.

  As shown in FIG. 6, the image processing apparatus according to one embodiment of the present invention includes four CPUs of a first core 10a-1, a second core 10a-2, a third core 10b-1, and a fourth core 10b-2. May be configured. Further, the surveillance camera according to one embodiment of the present invention may include an image processing device including the above four CPUs, like the surveillance camera 1a. In this case, the monitoring camera 1a includes a power supply unit 7 instead of the first and second power supply units 7a and 7b, and a memory 8 instead of the first and second memories 8a and 8b.

  The same processing as in the first image processing device 10a is performed by the first core 10a-1 and the second core 10a-2. The third core 10b-1 and the fourth core 10b-2 perform the same processing as the second image processing device 10b.

  The role sharing of the processing between the first core 10a-1 and the second core 10a-2 and the role sharing of the processing between the third core 10b-1 and the fourth core 10b-2 are set arbitrarily, respectively. Is also good. Further, the number of cores constituted by the CPU may be arbitrarily changed.

  In the present embodiment, the first image processing device 10a including the setting switching unit 11 and the monitoring target detection unit 12 is separate from the second image processing device 10b including the monitoring image processing execution unit 13. However, the present invention is not limited to this case. Although not shown, the setting switching unit 11, the monitoring target detection unit 12, and the monitoring image processing execution unit 13 may be provided in one image processing apparatus.

  Furthermore, in the present embodiment, the first and second image processing devices 10a and 10b are built in the monitoring camera 1, but the present invention is not limited to this case. For example, although not shown, the first and second image processing devices 10a and 10b may be built in the monitoring server 102, or may be built in an information processing device separate from the monitoring server 102 and the external storage device 103. Is also good. Alternatively, each of the first and second image processing apparatuses 10a and 10b may be a single unit.

  In these cases, the first and second image processing apparatuses 10a and 10b and the monitoring camera are wirelessly connected via the wireless LAN 101 or the like. In other words, the monitoring system according to one embodiment of the present invention only needs to include at least the image processing device and the monitoring camera connected (wired / wireless) to the image processing device.

[Embodiment 2]
Another embodiment of the present invention will be described below with reference to FIGS. 1, 2, 7, and 8. For convenience of explanation, members having the same functions as the members described in the above embodiment are denoted by the same reference numerals, and description thereof will be omitted.

<Functional configuration of surveillance system and surveillance camera>
First, the functional configuration of the monitoring system 200 and the monitoring camera 2 will be described with reference to FIGS. FIG. 1 is a block diagram illustrating an example of a functional configuration of the monitoring system 200. FIG. 2 is a block diagram illustrating an example of a functional configuration of the monitoring camera 2. FIG. 7 is a schematic diagram illustrating an example of setting a monitoring area of interest in the daytime.

  The surveillance system 200 shown in FIG. 1 includes a surveillance camera 2 instead of the surveillance camera 1. Other configurations and functions of the monitoring system 200 are the same as those of the monitoring system 100. The surveillance camera 2 shown in FIG. 2 includes a first image processing device 20a instead of the first image processing device 10a. Other configurations and functions of the monitoring camera 2 are the same as those of the monitoring camera 1.

  In the first image processing apparatus 20a illustrated in FIG. 2, when the setting switching unit 11 determines that one of the first determination criteria (any one of the plurality of determination criteria) is satisfied, attention monitoring in the recorded image R is performed. The area is switched from the first area F11 / F12 to the second area F21 / F22.

  When the setting switching unit 11 of the first image processing device 20a determines that one of the second determination criteria (any of the plurality of determination criteria) is satisfied, the setting switching unit 11 sets the attention monitoring area in the recorded image R to the second. Switching from the first area F11 to the second areas F21 and F22. The operation when it is determined that the other of the first and second setting switching conditions is satisfied is the same as that of the setting switching unit 11 of the first image processing apparatus 10a.

  When either of the first and second determination criteria is satisfied, it is necessary to switch the target monitoring area in the recorded image R. In the present embodiment, a first specific time at which the day is switched from daytime to nighttime in summer is one of the first determination criteria. In the present embodiment, 19:00 is set as the first specific time. In addition, the second specific time at which the day is switched from night to day in summer is the other of the first determination criteria. In the present embodiment, 9:00 is set as the second specific time.

  In the present embodiment, a third specific time at which the daytime is switched to the nighttime in a season other than summer is one of the second determination criteria. In the present embodiment, 18:00 is set as the third specific time. In addition, the fourth specific time at which the day is switched from night to day in a season other than summer is the other of the second determination criteria. In the present embodiment, 10:00 is set as the fourth specific time.

  Whether the season is summer or the season is other than summer is determined by date. In the present embodiment, the summer season is between June 1 and September 30, and the season other than summer is between October 1 and May 31.

  As shown in FIG. 7, when the season is summer, the target monitoring area is initially set in the first area F11 and the first area F13 in the recorded image R. The first area F11 is a recorded image portion including the location of the cash register in the recorded image R as in the first embodiment, and the first area F13 is a seasonal product (a product that sells well in summer) in the recorded image R. Is a recorded image portion that includes the display location of.

  On the other hand, when the season is other than the summer season, the attention monitoring area is initially set in the first area F11 in the recorded image R. In this case, since it is unlikely that a seasonal product (a product that sells well in summer) will be taken away, the recorded image portion including the display location of the seasonal product (a product that sells well in summer) in the recorded image R is deleted. Exclude from the first area.

  When the season is other than the summer season, a recorded image portion including a display location of another product in the recorded image R in addition to the first region F11 may be set as the first region. When the season is winter, the first area F13 may be replaced with a recorded image portion including a display location of a seasonal product (a product that sells well in winter) in the recorded image R.

  Further, similarly to the first embodiment, the second area F21 is a recorded image portion including the entrance for the employee in the recorded image R, and the second area F22 includes the entrance of the store in the recorded image R. This is a recorded image portion (see FIG. 3B).

<Image processing method by first image processing device and second image processing device>
Next, an image processing method by the first image processing device 20a and the second image processing device 10b will be described with reference to FIG. FIG. 8 is a flowchart illustrating an example of the image processing method.

  First, when the monitoring camera 2 starts the recording process of the entire monitoring area 800, the process of the flowchart shown in FIG. 8 is started (START). In S201, the monitoring image processing execution unit 13 initializes the first area F11 / F13 in the recorded image R (when the season is summer) or the first area F11 in the recorded image R (when the season is other than summer). The recording process of the set attention monitoring area is started, and the process proceeds to S202.

  Next, the setting switching unit 11 checks the date in S202, checks the time in S203, and proceeds to S204. In S204, based on the confirmed date and time, the setting switching unit 11 determines whether one of the first criteria or one of the second criteria is satisfied (whether 19:00 is reached in summer, Alternatively, it is determined whether or not 18:00 has been reached in a season other than summer. When it is determined as N in S204, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the first area F11 / F13 or the attention monitoring area initially set in the first area F11 ( S206: Monitoring target detection step).

  On the other hand, when it is determined in S204 that one of the first determination criteria is satisfied (determined as Y), the setting switching unit 11 changes the target monitoring area in the recorded image R from the first areas F11 and F13 in S205. After switching to the two areas F21 and F22, the process proceeds to S206. Alternatively, when it is determined in S204 that one of the second determination criteria is satisfied (determined as Y), the setting switching unit 11 changes the target monitoring area in the recorded image R from the first area F11 to the second area in S205. After switching to F21 / F22, the process proceeds to S206.

  Next, in S206, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the target monitoring region switched to the second region F21 / F22 (S206: monitoring target detection step).

  The processes in S207 to S209 when Y is determined in S206 are the same as the processes in S106 to S108 in the flowchart shown in FIG. In addition, since each process after the determination of N in S206 is the same as the flowchart shown in FIG. 5, the description thereof will be omitted.

  After execution of the process of S209, the process returns to S202. The setting switching unit 11 checks the date in S202, checks the time in S203, and proceeds to S204. In S204, the setting switching unit 11 determines whether one of the other of the first criterion and the other of the second criterion is satisfied based on the confirmed date and time (whether 9:00 is reached in summer, Alternatively, it is determined whether 10:00 has been reached in a season other than summer. If it is determined as N in S204, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring area switched to the second area F21 or F22 in S206.

  On the other hand, when it is determined in S204 that the other of the first determination criteria is satisfied (determined as Y), the setting switching unit 11 changes the attention monitoring area in the recorded image R from the second areas F21 and F22 to S20 in S205. The process returns to the one area F11 / F13 and proceeds to S206. Alternatively, when it is determined in S204 that the other of the second determination criteria has been satisfied (determined as Y), the setting switching unit 11 changes the target monitoring area in the recorded image R from the second areas F21 and F22 to the second area in S205. The process returns to the one area F11 and proceeds to S206.

  Next, in S206, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the first area F11 / F13 or the attention monitoring area set again in the first area F11. The processes in S207 to S209 are the same as above, and the processes in S202 to S209 are continuously executed until the monitoring of the monitoring area 800 by the monitoring system 200 ends.

[Embodiment 3]
Another embodiment of the present invention will be described below with reference to FIGS. 1, 2, and 9. For convenience of explanation, members having the same functions as the members described in the above embodiment are denoted by the same reference numerals, and description thereof will be omitted.

<Functional configuration of surveillance system and surveillance camera>
First, a functional configuration of the monitoring system 300 and the monitoring camera 3 will be described with reference to FIGS. FIG. 1 is a block diagram illustrating an example of a functional configuration of the monitoring system 300. FIG. 2 is a block diagram illustrating an example of a functional configuration of the monitoring camera 3.

  The surveillance system 300 shown in FIG. 1 includes a surveillance camera 3 instead of the surveillance cameras 1 and 2. Other configurations and functions of the monitoring system 300 are the same as those of the monitoring systems 100 and 200. The surveillance camera 3 shown in FIG. 2 includes a first image processing device 30a instead of the first image processing devices 10a and 20a. Other configurations and functions of the monitoring camera 3 are the same as those of the monitoring cameras 1 and 2.

  In the first image processing device 30a illustrated in FIG. 2, the setting switching unit 11 determines whether the switching determination stop condition (second condition) is satisfied. If it is determined that the switching determination suspension condition has been satisfied, the setting switching unit 11 of the first image processing device 30a sets the target monitoring areas initially set in the first areas F11 and F12 to the second areas F21 and F22 (the third areas F21 and F22). ) (See FIG. 3). The operation when it is determined that the switching determination suspension condition is not satisfied is the same as that of the setting switching unit 11 of the first image processing apparatuses 10a and 20a.

  When the switching determination stop condition is satisfied, the setting switching unit 11 of the first image processing device 30a stops determining whether the switching condition is satisfied. In the present embodiment, the switchover determination stop condition is that the store is not on an operating day. In this embodiment, holidays and regular holidays (Monday) are set as non-operating days of the store. Whether or not it is a store operation day is determined based on the date and the day of the week.

  As described above, the switching determination stop condition that the store is not a working day is that when there is no store working day, no customers come and go over the whole day, and the store situation approximates the nighttime situation on the work day. Therefore, fixing the attention monitoring area to the second areas F21 and F22 is in accordance with the actual situation of the store.

  It should be noted that the above-described fixing of the monitoring area of interest to the second areas F21 and F22 is merely an example, and is not limited to this case. According to the content of the switching determination suspension condition, the target monitoring area initially set in the first areas F11 and F12 may be fixed to the first areas F11 and F12 as it is. Alternatively, the target monitoring area may be fixed to a specific area (third area) different from the first areas F11 and F12 and the second areas F21 and F22.

  The monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring area fixed to the second areas F21 and F22.

<Image processing method by first image processing device and second image processing device>
Next, an image processing method by the first image processing device 30a and the second image processing device 10b will be described with reference to FIG. FIG. 9 is a flowchart illustrating an example of the image processing method.

  First, when the monitoring camera 3 starts the recording process of the entire monitoring area 800, the process of the flowchart shown in FIG. 9 is started (START). In S301, the monitoring image processing execution unit 13 starts the recording processing of the target monitoring area initially set in the first areas F11 and F12 in the recorded image R, and proceeds to S302.

  Next, the setting switching unit 11 checks the date in S302, checks the day of the week in S303, and proceeds to S304. In S304, based on the confirmed date and day of the week, the setting switching unit 11 determines whether or not the switching determination suspension condition is satisfied (is not a store operation day). When it is determined to be N in S304, the setting switching unit 11 checks the time in S305, and proceeds to S306.

  In S306, the setting switching unit 11 determines whether the confirmed time satisfies one of the switching conditions (whether 18:30). When it is determined to be N in S306, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring region initially set in the first regions F11 and F12 (S309: monitoring target detection step). ).

  On the other hand, if it is determined as Y in S306, the setting switching unit 11 switches the attention monitoring area in the recorded image R from the first areas F11 and F12 to the second areas F21 and F22 in S307, and proceeds to S309. In S309, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring area switched to the second areas F21 and F22.

  If Y is determined in S304, the setting switching unit 11 fixes the target monitoring area to the second areas F21 and F22 in S308, and proceeds to S309. In S309, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring area fixed to the second areas F21 and F22.

  The processing in S310 to S312 when Y is determined in S309 is the same as the processing in S106 to S108 in the flowchart shown in FIG. 5, and a description thereof will be omitted. In addition, since each process after the determination of N in S309 is the same as the flowchart shown in FIG. 5, the description thereof will be omitted.

  After performing the process of S312, the process returns to S302. Since each processing of S302 to S305 and S308 is as described above, the description thereof is omitted. In S306, the setting switching unit 11 determines whether the confirmed time satisfies the other of the switching conditions (whether 9:30). If it is determined to be N in S306, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the target monitoring area that has been switched to the second areas F21 and F22 in S309.

  On the other hand, if it is determined as Y in S306, the setting switching unit 11 returns the attention monitoring area in the recorded image R from the second areas F21 and F22 to the first areas F11 and F12 in S307, and proceeds to S309. Next, in S309, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring area set again in the first areas F11 and F12. The processes in S310 to S312 are the same as described above, and the processes in S302 to S312 are continuously executed until the monitoring of the monitoring area 800 by the monitoring system 300 ends.

[Embodiment 4]
Another embodiment of the present invention will be described below with reference to FIGS. 1, 2, and 10. For convenience of explanation, members having the same functions as the members described in the above embodiment are denoted by the same reference numerals, and description thereof will be omitted.

<Functional configuration of surveillance system and surveillance camera>
First, a functional configuration of the monitoring system 400 and the monitoring camera 4 will be described with reference to FIGS. FIG. 1 is a block diagram illustrating an example of a functional configuration of the monitoring system 400. FIG. 2 is a block diagram illustrating an example of a functional configuration of the monitoring camera 4.

  The surveillance system 400 shown in FIG. 1 includes a surveillance camera 4 instead of the surveillance cameras 1, 2, and 3. Other configurations and functions of the monitoring system 400 are the same as those of the monitoring systems 100, 200, and 300. The surveillance camera 4 shown in FIG. 2 includes a first image processing device 40a instead of the first image processing devices 10a, 20a, and 30a. Other configurations and functions of the monitoring camera 4 are the same as those of the monitoring cameras 1, 2, and 3.

  In the first image processing device 40a illustrated in FIG. 2, the setting switching unit 11 determines whether the setting release condition (third condition) is satisfied after switching from the first region of the attention monitoring region to the second region. Then, when it is determined that the setting release condition is satisfied, the setting of the attention monitoring area is released. The operation when it is determined that the setting release condition is not satisfied is the same as that of the setting switching unit 11 of the first image processing apparatuses 10a, 20a, and 30a.

  The setting release condition is a condition for releasing the setting of the target monitoring area switched to the second area in the recorded image. Depending on which monitoring area the monitoring system 400 monitors, various setting cancellation conditions can be set. In addition, various regions (recorded image portions) in the recorded image can be set as the second region according to the type of the location serving as the monitoring area.

  For example, if the monitoring area is a store displaying rain gear supplies, it is conceivable to set the recorded image portion including the display location of the rain gear supplies in the recorded image in the second area. In this case, if the switching condition is that rain has started to fall and the setting release condition is that rain has stopped, a suspicious person 900 (see FIG. 4) attempting to bring out rain gear can be efficiently detected. .

  Also, for example, if the monitoring area is a store displaying new products, it is conceivable to set a recorded image portion including the display location of the new product in the recorded image in the second area. In this case, it is conceivable that the switching condition is set immediately after a new product is received, and the setting release condition is set such that the new product is sold out.

  Further, for example, if the monitoring area is a ticket gate of a station, it is conceivable to set a recorded image portion including an automatic ticket gate in the recorded image in the second area. In this case, the switching condition may be that the user has purchased the ticket at the automatic ticket vending machine, and the setting release condition may be that the user has passed the automatic ticket gate. Furthermore, if the monitoring area is a platform of a station, it is conceivable to set a recorded image portion including a train door in the recorded image in the second area. In this case, it is conceivable that the arrival of the train is used as the switching condition, and the departure of the train is used as the setting release condition.

  Further, for example, if the monitoring area is a parking lot, it is conceivable to set the recorded image portion including the entrance of the parking lot in the recorded image in the second area. In this case, the switching condition may be that the user has taken the parking ticket, and the setting release condition may be that the vehicle of the user has left the entrance of the parking lot. Alternatively, it is conceivable to set the recorded image portion including the exit of the parking lot in the recorded image in the second area. In this case, it is conceivable that the switching condition is that the user has completed the payment using the parking fee payment machine, and the setting release condition is that the vehicle of the user has left the exit of the parking lot.

  In each of the above examples, the event that is the content of the switching condition and the event that is the content of the setting release condition are generally referred to as “switching elements”. The information on the “switching element” is received by the setting switching unit 11 from an external communication device or the like via the wireless LAN 101, for example.

<Image processing method by first image processing device and second image processing device>
Next, an image processing method by the first image processing device 40a and the second image processing device 10b will be described with reference to FIG. FIG. 10 is a flowchart illustrating an example of the image processing method.

  First, when the monitoring camera 4 starts the recording process of the entire monitoring area, the process of the flowchart shown in FIG. 10 is started (START). In S401, the monitoring image processing execution unit 13 starts the recording process of the target monitoring area initially set in the first area in the recorded image, and proceeds to S402.

  In S402, the setting switching unit 11 confirms the switching element, and proceeds to S403. In S403, the setting switching unit 11 determines whether the switching condition is satisfied based on the confirmed switching element. When it is determined to be N in S403, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the target monitoring area initially set in the first area (S405: monitoring target detection step).

  On the other hand, if Y is determined in S403, the setting switching unit 11 switches the target monitoring area in the recorded image from the first area to the second area in S404, and proceeds to S405. In S405, the monitoring target detection unit 12 detects whether or not the suspicious individual 900 exists in the attention monitoring area switched to the second area.

  Each process of S406 to S408 when Y is determined in S405 is the same as each process of S106 to S108 in the flowchart shown in FIG. 5, and therefore the description thereof is omitted. In addition, since each process after the determination of N in S405 is the same as the flowchart shown in FIG. 5, the description thereof will be omitted.

  After executing the processing of S408, the process returns to S402. The setting switching unit 11 checks the switching element in S402, and determines whether the setting release condition is satisfied based on the checked switching element in S403. If it is determined to be N in S403, the monitoring target detection unit 12 detects in S405 whether or not the suspicious individual 900 exists in the attention monitoring area that has been switched to the second area.

  The processes in S406 to S408 are the same as described above, and the processes in S402 to S408 are continuously executed until the monitoring of the monitoring area 800 by the monitoring system 400 ends. On the other hand, if it is determined as Y in S403, the setting switching unit 11 cancels the setting of the target monitoring area in the recorded image in S404.

  In each of the monitoring cameras 1 to 4 of the first to fourth embodiments shown in FIG. 2, the first power supply unit 7a and the second power supply unit 7b, and the first memory 8a and the second memory 8b are separately installed. However, these may be configured to also serve as one power supply unit 7 or one memory 8 as shown in FIG.

[Example of software implementation]
The control blocks (especially the monitoring target detection unit 12 and the monitoring image processing execution unit 13) of the first image processing devices 10a to 40a and the second image processing device 10b include logic circuits (IC chips) formed on an integrated circuit (IC chip) or the like. Hardware) or software using a CPU (Central Processing Unit).

  In the latter case, the first image processing apparatuses 10a to 40a and the second image processing apparatus 10b are configured to execute a command of a program, which is software for realizing each function, and to execute the program and various data by a computer (or CPU). A ROM (Read Only Memory) or a storage device (these are referred to as “recording media”) recorded in a readable manner, a RAM (Random Access Memory) for expanding the above-described program, and the like are provided. Then, the object of the present invention is achieved when the computer (or CPU) reads the program from the recording medium and executes the program. As the recording medium, a “temporary tangible medium” such as a tape, a disk, a card, a semiconductor memory, or a programmable logic circuit can be used. Further, the program may be supplied to the computer via an arbitrary transmission medium (a communication network, a broadcast wave, or the like) capable of transmitting the program. Note that one embodiment of the present invention can also be realized in the form of a data signal embedded in a carrier wave, in which the program is embodied by electronic transmission.

[Summary]
The image processing apparatuses (first image processing apparatuses 10a, 20a, 30a, 40a, and second image processing apparatus 10b) according to the first aspect of the present invention include recorded images recorded by surveillance cameras (1, 2, 3, and 4). An image processing apparatus for processing (R), wherein a monitoring target (suspicious person 900) exists in a target monitoring area set in a first area (F11, F12, F13) in the recorded image. A monitoring target detection unit (12) for detecting whether the monitoring target has been detected, and a monitoring image processing execution unit (13) for performing monitoring image processing for monitoring the monitoring target with respect to the image in the target monitoring region (target monitoring region image I). The monitoring image processing execution unit executes the monitoring image processing when the monitoring target detection unit detects the presence of the monitoring target.

  According to the configuration, the monitoring target detection unit detects whether or not the monitoring target exists in the target monitoring area set in the recorded image recorded by the monitoring camera. The monitoring image processing execution unit executes the monitoring image processing when the monitoring target detection unit detects the presence of the monitoring target.

  Therefore, for example, by setting the recording target of the monitoring camera to the entire monitoring area, the image processing apparatus is switched to the high power consumption mode only while the monitoring target is present in the target monitoring area, and the image in the area is monitored. , Monitoring image processing can be continued at a high resolution. Further, this processing can be performed by one monitoring camera.

  Furthermore, since the operation of the monitoring target in the entire monitoring area can be recorded and image-processed, it is possible to know where the monitoring target has entered and escaped from. Therefore, the monitoring accuracy can be improved while suppressing the calculation cost and reducing the cost.

  In the image processing device according to the second aspect of the present invention, in the first aspect, the monitoring image processing execution unit detects the presence of the monitoring target by the monitoring target detection unit after starting execution of the monitoring image processing. When the monitoring image processing runs out, the execution of the monitoring image processing may be stopped.

  According to the above configuration, the monitoring image processing execution unit stops the execution of the monitoring image processing when the monitoring target detection unit stops detecting the presence of the monitoring target after the execution of the monitoring image processing is started. Therefore, for example, when the monitoring target no longer exists, the image processing apparatus can be switched to the low power consumption mode, and the recording and image processing of the entire monitoring area can be performed at a low resolution. Therefore, it is possible to further suppress the calculation cost and reduce the cost while improving the monitoring accuracy.

  The image processing device according to an aspect 3 of the present invention, according to the aspect 1 or 2, determines whether a first condition that requires switching of the attention monitoring area in the recorded image is satisfied, and When it is determined that the condition is satisfied, the setting switching unit that switches the attention monitoring area in the recorded image from the first area to the second area (F21, F22) different from the first area (F11, F12). (11) The monitoring target detection unit may further detect whether or not the monitoring target exists in the target monitoring region set in the second region.

  According to the above configuration, the image processing device according to an aspect of the present invention further includes a setting switching unit that switches the target monitoring area from the first area to the second area when determining that the first condition is satisfied. ing. Further, the monitoring target detection unit detects whether or not the monitoring target exists in the target monitoring region set in the second region.

  Therefore, for example, when a location in the monitoring area where the monitoring target is likely to appear and disappear changes in accordance with a status change inside and outside the monitoring area, the target monitoring area set before the status change is set to another area (second area) in the recorded image. ), It is possible to appropriately respond to this situation change. Therefore, the monitoring accuracy can be further improved while suppressing the calculation cost and reducing the cost.

  In the image processing device according to the fourth aspect of the present invention (first image processing device 20a), in the third aspect, the first condition includes a plurality of determination criteria, and the setting switching unit includes a plurality of the determination criteria. May be switched from the first area (F11, F13) to the second area in the recorded image.

  For example, if the monitored area is a commercial facility such as a store, the main time zone during which customers enter and leave differs depending on the season. As described above, depending on the situation change inside and outside the monitoring area, it may be preferable to change the condition that requires the switching of the target monitoring area. In this regard, according to the above configuration, the setting switching unit switches the target monitoring area from the first area to the second area when it is determined that any of the plurality of determination criteria is satisfied.

  Therefore, by setting a plurality of determination criteria corresponding to a change in the situation inside and outside the monitoring area, even when the situation inside and outside the monitoring area changes in various forms, the target monitoring area can be appropriately switched. Therefore, the monitoring accuracy can be further improved while suppressing the calculation cost and reducing the cost.

  In the image processing device (first image processing device 30a) according to aspect 5 of the present invention, in the above-described aspect 3, the setting switching unit may determine whether a second condition for canceling the satisfaction determination of the first condition is satisfied. Determining whether or not the second condition is satisfied, fixing the target monitoring area to either the first area or a third area (F21, F22) different from the first area; The monitoring target detection unit may detect whether or not the monitoring target exists in the target monitoring region fixed in one of the first region and the third region.

  According to the above configuration, when the setting switching unit determines that the second condition is satisfied, the setting switching unit fixes the target monitoring region to either the first region or the third region. In addition, the monitoring target detection unit detects whether or not the monitoring target exists in the target monitoring region fixed to either the first region or the third region by the setting switching unit.

  Therefore, when it becomes unnecessary to switch the attention monitoring area in the recorded image from the first area to the second area due to a change in the situation inside and outside the monitoring area, the monitoring target area is changed to a specific area ( (The first region or the third region). Therefore, useless switching of the monitoring target area can be prevented, and the monitoring accuracy can be improved, and the calculation cost can be further reduced and the cost can be reduced.

  An image processing device (first image processing device 40a) according to an aspect 6 of the present invention, in any one of the aspects 3 to 5, wherein the setting switching unit is configured to switch the first area of the attention monitoring area from the second area to the second area. After switching to, it is determined whether or not a third condition for canceling the setting of the attention monitoring area in the recorded image is satisfied. If it is determined that the third condition is satisfied, the attention monitoring is performed. The setting of the area may be canceled.

  According to the above configuration, when the setting switching unit determines that the third condition is satisfied, the setting switching unit cancels the setting of the target monitoring area switched to the second area. Therefore, by appropriately setting the contents of the third condition, after the attention monitoring area is switched to the second area, the setting of the attention monitoring area is continued while the attention monitoring area is required, and the attention monitoring is continued. When the monitoring of the area is no longer necessary, the setting can be canceled. Therefore, it is possible to prevent useless monitoring of the monitoring target area, and to further reduce the calculation cost and reduce the cost while improving the monitoring accuracy.

  The surveillance camera (1, 2, 3, 4) according to an aspect 7 of the present invention provides the image processing apparatus (the first image processing apparatus 10a, 20a, 30a, 40a, Processing device 10b). According to the above configuration, it is possible to realize a surveillance camera capable of improving the monitoring accuracy while suppressing the calculation cost and reducing the cost.

  A surveillance system (100, 200, 300, 400) according to an aspect 8 of the present invention includes a surveillance camera connected to the image processing device according to any one of the aspects 1 to 6. According to the above configuration, it is possible to realize a monitoring system capable of improving monitoring accuracy while suppressing calculation costs and reducing costs.

  An image processing method according to a ninth aspect of the present invention is an image processing method for processing a recorded image recorded by a surveillance camera, wherein attention is set to a first area (F11, F12, F13) in the recorded image. A monitoring target detection step (S105, S107, S206, S208, S309, S311, S405, S407) for detecting whether or not a monitoring target exists in the monitoring region; A monitoring image processing execution step of performing monitoring image processing for monitoring a target (S106, S207, S310, S406), wherein when the presence of the monitoring target is detected in the monitoring target detection step, In the monitoring image processing execution step, the monitoring image processing is performed.

  According to the above configuration, it is possible to realize an image processing method capable of improving monitoring accuracy while suppressing and reducing calculation costs.

  The image processing apparatus according to each aspect of the present invention may be realized by a computer. In this case, the computer is operated as each unit (software element) included in the image processing apparatus, so that the image processing apparatus is provided to the computer. The present invention also includes a program for an image processing apparatus to be realized by a computer and a computer-readable recording medium on which the program is recorded.

[Appendix]
The present invention is not limited to the embodiments described above, and various modifications are possible within the scope shown in the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. Is also included in the technical scope of the present invention. Furthermore, new technical features can be formed by combining the technical means disclosed in each embodiment.

1, 1a, 2, 3, 4 Surveillance cameras 10a, 20a, 30a, 40a First image processing device (image processing device)
10b Second image processing device (image processing device)
Reference Signs List 11 setting switching unit 12 monitoring target detection unit 13 monitoring image processing execution unit 100, 200, 300, 400 monitoring system 900 suspicious person (monitoring target)
F11, F12, F13 First area F21, F22 Second area I Attention monitoring area image (image in the attention monitoring area)
R Recorded image

Claims (10)

An image processing device that processes a recorded image recorded by a surveillance camera,
A monitoring target detection unit that detects whether or not a monitoring target exists in a target monitoring region set in a first region in the recorded image;
A monitoring image processing execution unit that performs monitoring image processing for monitoring the monitoring target with respect to the image in the attention monitoring area,
The image processing apparatus, wherein the monitoring image processing execution unit executes the monitoring image processing when the monitoring target detection unit detects the presence of the monitoring target.   The monitoring image processing execution unit, after starting the execution of the monitoring image processing, stops the execution of the monitoring image processing when the monitoring target detection unit stops detecting the presence of the monitoring target. The image processing apparatus according to claim 1. It is determined whether a first condition that requires switching of the attention monitoring area in the recorded image is satisfied. If it is determined that the first condition is satisfied, the attention monitoring area in the recorded image is determined. A setting switching unit that switches from the first area to a second area different from the first area,
The image processing apparatus according to claim 1, wherein the monitoring target detection unit detects whether or not the monitoring target exists in the target monitoring region set in the second region. . The first condition includes a plurality of criteria,
The setting switching unit switches the monitoring area of interest in the recorded image from the first area to the second area when it is determined that any one of the plurality of determination criteria is satisfied. The image processing device according to claim 3. The setting switching unit determines whether a second condition for canceling the satisfaction determination of the first condition is satisfied. When determining that the second condition is satisfied, the setting switching unit sets the attention monitoring area to the predetermined condition. Fixed to either the first region or a third region different from the first region,
4. The monitoring target detection unit detects whether the monitoring target is present in the target monitoring region fixed in one of the first region and the third region. An image processing apparatus according to claim 1.   The setting switching unit determines whether a third condition for canceling the setting of the attention monitoring area in the recorded image is satisfied after the attention monitoring area is switched from the first area to the second area. 6. The image processing apparatus according to claim 3, wherein when it is determined that the third condition is satisfied, the setting of the attention monitoring area is released.   A surveillance camera comprising the image processing device according to claim 1.   A surveillance system, comprising: a surveillance camera connected to the image processing apparatus according to claim 1.   A program for causing a computer to function as the image processing apparatus according to claim 1, wherein the program causes a computer to function as the monitoring target detection unit and the monitoring image processing execution unit. An image processing method for processing a recorded image recorded by a surveillance camera, the method comprising: detecting a monitoring target in a monitoring target area set in a first area in the recorded image; Steps and
A monitoring image processing execution step of performing monitoring image processing for monitoring the monitoring target for an image in the attention monitoring area,
An image processing method, wherein when the presence of the monitoring target is detected in the monitoring target detection step, the monitoring image processing is executed in the monitoring image processing execution step.

Images