Image processing system, image processing method, and computer program

Abstract

To provide an image processing system capable of facilitating setting of a threshold of behavior detection of a human body.SOLUTION: An image processing system includes: acquisition means for acquiring a photographed image photographed by an imaging device; detection means for, based on a motion of a human body to be obtained from the photographed image, detecting each one of predetermined behaviors of the human body; determination means for determining presence / absence of a suspicious behavior of the human body based on a detection result of the detection means; warning means for issuing a warning if the determination means determines presence of the suspicious behavior; setting means for setting a detection threshold for each behavior in the detection means; and testing means for testing propriety of detection by the detection means for each behavior.SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to an image processing system, an image processing method, and a computer program capable of detecting behavior

Background Art

[0002] In recent years, systems have been developed that analyze video captured by surveillance cameras to detect suspicious behavior from specific behaviors of the human body or a combination of a plurality of behaviors. Although the behaviors to be detected differ depending on the system, for example, for the purpose of suppressing shoplifting in retail stores and the like, there exists a technology that detects suspicious behavior from a combination of behaviors such as the trajectory of the human body and head turning, and issues a warning.

[0003] Patent Document 1 relates to a technology for speaking to a human body exhibiting suspicious behavior using a directional speaker. Here, as an example, a method is disclosed for determining that an action is suspicious when the variance of the face orientation exceeds a threshold, or when the cumulative time during which the face orientation is not in the movement direction exceeds a threshold.

[0004] Patent Document 2 relates to a technology for detecting suspicious behavior of the human body. Here, a method is disclosed for determining that an action is suspicious when the time-series variation amount of the face orientation exceeds a threshold, or when the human body is looking in a unobstructed direction that can be seen through.

Prior Art Documents

Patent Documents

[0005]

Patent Document 1

Patent Document 2

Summary of Invention

Problems to be Solved by the Invention

[0006] Both the technology described in Patent Document 1 and the technology described in Patent Document 2 require setting a threshold for detecting human body movements in order to determine suspicious behavior. A reasonable value for this threshold will vary depending on the field of view of the surveillance camera and the installation environment, and adjustment may be necessary. However, Patent Document 1 does not mention a method for setting the threshold or a method for confirming whether each behavior can be detected using actual video footage.

[0007] Furthermore, while Patent Document 2 states that threshold settings may be determined by conducting preliminary experiments or observing human behavior captured by surveillance cameras, it does not mention specific setting methods, including the user interface. Therefore, if a suspicious behavior detection system is activated but does not perform detection as intended, it may be difficult to determine which behavior threshold setting was problematic.

[0008] In view of the aforementioned problems, the present invention aims to provide an image processing system that allows for easy setting of a threshold for detecting human body behavior. [Means for solving the problem]

[0009] One aspect of the image processing system of the present invention is An acquisition means for acquiring captured images taken by an imaging device, A detection means for detecting a predetermined number of human body movements based on the human body movements obtained from the captured image, A determination means for determining whether or not there is suspicious behavior of the human body based on the detection result by the detection means, If the aforementioned determination means determines that the aforementioned suspicious behavior has occurred, a warning means provides a warning, A setting means for setting a detection threshold for each behavior in the detection means, The system is characterized by having a test means for testing whether the detection means can detect each of the aforementioned behaviors. [Effects of the Invention]

[0010] According to the present invention, an image processing system can be realized that allows for easy setting of the threshold for detecting human body behavior. [Brief explanation of the drawing]

[0011] [Figure 1] This is a functional block diagram showing an example configuration of the image processing system 100 according to Embodiment 1. [Figure 2] This is a block diagram showing an example of the hardware configuration of the image processing device 101 according to Embodiment 1. [Figure 3] This diagram shows the state transitions of the image processing device 101 according to Embodiment 1. [Figure 4] This flowchart shows an example of operation using the UI device 103 according to Embodiment 1. [Figure 5] This is a flowchart that continues from Figure 4. [Figure 6] (A) is a diagram showing an example of the settings screen in the UI device 103 according to Embodiment 1, and (B) is a diagram showing an example of the settings screen in the UI device 103 according to Embodiment 2. [Figure 7] Figures (A) and (B) show examples of confirmation screens for whether behavior can be detected by the UI device 103. [Figure 8] This is a flowchart showing an example of operation using the UI device 103 according to Embodiment 2. [Figure 9] This is a flowchart that continues from Figure 8. [Modes for carrying out the invention]

[0012] Embodiments of the present invention will be described below with reference to the drawings. However, the present invention is not limited to the following embodiments. In each drawing, the same reference numeral is used for the same member or element, and redundant explanations are omitted or simplified.

[0013] (Embodiment 1) First, an example of the configuration of an image processing system 100 corresponding to an embodiment will be described with reference to Figure 1. Figure 1 is a functional block diagram illustrating a configuration example of the image processing system 100 according to the first embodiment. Note that part of the functional blocks shown in Figure 1 is implemented by causing a CPU 201 as a computer included in the image processing apparatus to execute a computer program stored in a memory as a storage medium.

[0014] However, part or all of them may be implemented by hardware. As the hardware, a dedicated circuit (ASIC), a processor (reconfigurable processor, DSP), or the like can be used. Further, each of the functional blocks shown in Figure 1 does not need to be housed in the same casing, and may be configured by separate apparatuses connected to each other via a signal path.

[0015] The image processing system 100 is configured by connecting an image processing apparatus 101, an imaging apparatus 102, a UI apparatus (user interface apparatus) 103, and a warning apparatus 104 via a network 208 such as the Internet which will be described later. The image processing apparatus 101 is, for example, a PC (personal computer). The imaging apparatus 102 is, for example, a network camera, and may be one or a plurality of units.

[0016] An image acquisition unit 105 acquires an image from the imaging apparatus 102. Here, the image acquisition unit 105 directly acquires the image from the imaging apparatus 102, but the image acquisition unit 105 may acquire a video from a recorded image stored in an HDD 204 which will be described later with reference to Figure 2.

[0017] Note that the image acquisition unit 105 functions as an acquisition unit that acquires a captured image captured by the imaging apparatus, and as described above, can acquire a live video captured by the imaging apparatus or a video pre-stored in a storage device as the captured image.

[0018] The behavior detection unit 106 functions as a detection means and detects a predetermined set of behaviors based on the movements of the human body obtained from the captured images acquired by the image acquisition unit 105. These behaviors are used to determine suspicious behavior, as described later, and include, but are not limited to, head shaking. As for methods of detecting behavior, for example, in the case of head shaking, one method is to detect the orientation of the human body and head using known image analysis techniques and estimate the head shaking situation from the change in orientation.

[0019] The suspicious behavior determination unit 107 functions as a determination means and performs a determination step to determine whether or not there is suspicious behavior of the human body based on the detection results of the behavior detected by the behavior detection unit 106.

[0020] The types of behaviors used to determine suspicious behavior may be independent of each other or a combination of multiple behaviors. Furthermore, the rules for determining suspicious behavior based on a specific behavior or combination of behaviors are either internally maintained by the image processing device 101 or set by the UI device 103.

[0021] The warning instruction unit 108 functions as a warning means, and if the suspicious behavior determination unit 107 determines that there is suspicious behavior by a person, it instructs the warning device 104 to issue a warning. The warning device 104 executes a warning step that warns the user by some means, such as a speaker, an alarm lamp, or a warning message displayed on the screen. The communication unit 109 receives instructions from the UI device 103 and also notifies the UI device 103.

[0022] The UI device 103 can be a notebook PC (notebook personal computer) with a display function, a portable smartphone or tablet PC, or a remote controller. By operating the UI device 103, the user can change the settings used in the image acquisition unit 105, the behavior detection unit 106, and the suspicious behavior determination unit 107, as well as the state of the image processing device 101.

[0023] Various screens (user interface) related to operation are displayed on the display unit of the UI device 103. An example of the display when changing the settings used by the behavior detection unit 106 by operating the UI device 103 will be described later using Figure 6. An example of checking whether behavior based on thresholds can be detected when the aforementioned settings are changed will be described later using Figure 7.

[0024] The changes in the state of the image processing device 101 will be described later using Figure 3. Note that the UI device 103 may also directly acquire and display images from the imaging device 102 when changing the settings of the image processing device 101.

[0025] Next, Figure 2 is a block diagram showing an example of the hardware configuration of the image processing device 101 according to Embodiment 1, and the example of the hardware configuration of the image processing device 101 according to Embodiment 1 will be explained using Figure 2. The image processing device 101 is an example of an information processing device and includes a CPU 201, ROM 202, RAM 203, HDD (hard disk drive) 204, communication unit 205, input device 206, and display device 207 (display means).

[0026] The CPU 201, acting as a computer, reads control computer programs recorded in the ROM 202, which acts as a storage medium, and executes various processes as shown in the flowcharts in Figures 4, 5, 8, and 9. The RAM 203 is used as a temporary storage area, such as the main memory or work area.

[0027] HDD204 is a storage device for saving image data and other data. The communication unit 205 is connected to the network 208 and is a circuit that performs communication processing with the imaging device 102, warning device 104, etc. via the network 208.

[0028] The input device 206 has a device for the user to manually configure, such as a keyboard or mouse. The display device 207 is a device such as a display that shows images and settings. The display device 207 may be located outside the image processing device 101, or it may be located in the UI device 103.

[0029] As described above, the hardware configuration of the image processing device 101 has hardware components similar to those found in a personal computer (PC). Therefore, the various functions realized by the image processing device 101 can be implemented as software that runs on a PC. The image processing device 101 can realize the various functional blocks shown in Figure 1 by having the CPU 201 execute a program.

[0030] The hardware configuration of the UI device 103 is almost the same as that shown in Figure 2. However, the communication unit 205 is a wireless LAN interface, and the network 208 is a wireless LAN. In addition, the input device 206 includes a touch panel.

[0031] Furthermore, the functions of the image processing device 101 and the warning device 104 may be built into the UI device 103. Alternatively, the functions of the UI device 103 and the warning device 104 may be built into the image processing device 101. In other words, although the image processing device 101 and the UI device 103 are described as separate components in Embodiment 1, they may be configured as an integrated image processing system. In that case, some or all of the configuration shown in Figure 2 may be common to the image processing device 101 and the UI device 103.

[0032] Figure 3 is a diagram showing the state transitions of the image processing device 101 according to Embodiment 1, and the state transitions in the image processing device 101 will be explained using Figure 3. The image processing device 101 has an operating state 301 and a warning stop state 302. The operating state refers to a state in which behavior detection and suspicious behavior determination are performed based on settings, and a warning instruction is given to the warning device 104 when suspicious behavior is determined.

[0033] The warning-stopped state is a state that mainly occurs during configuration changes, where behavior detection and suspicious behavior detection are performed as needed during the configuration change process, but no warning instructions are issued to the warning device 104. The warning-stopped state also includes a state in which warning sounds and warning displays are suppressed to be less conspicuous compared to the operating state 301. The transition from the operating state to the warning-stopped state, and the transition from the warning-stopped state to the operating state, can be switched by operating the UI device 103.

[0034] Figure 4 is a flowchart showing an example of operation in the UI device 103 according to Embodiment 1, and Figure 5 is a flowchart that continues from Figure 4. Using the flowcharts in Figures 4 and 5, we will explain, as an example of the operation procedure for the UI device 103, how to set the threshold for behavior detection in Embodiment 1 and how to confirm whether behavior can be detected at the set threshold using video.

[0035] Furthermore, the CPU, acting as a computer within the image processing system 100, executes the computer program stored in memory, thereby performing the actions shown in each step of the flowcharts in Figures 4 and 5. In the following explanation, the method of confirming detection feasibility using video will be referred to as a test. In this embodiment, it is assumed that one or more imaging devices are already registered with the image processing system 100.

[0036] First, in step S401, the CPU in the image processing system 100 checks whether the image processing device 101 is operational. If the image processing device 101 is operational, in step S402 the image processing device 101 is switched to a warning stop state before proceeding to step S403.

[0037] In other words, when performing the test, the warnings from the warning means are suppressed. If the answer in step S401 is No, proceed to step S403. In step S403, the UI device 103 displays the threshold setting screen used by the behavior detection unit 106.

[0038] Figure 6(A) shows an example of a setting screen in the UI device 103 according to Embodiment 1, and an example of the threshold setting screen display in Embodiment 1 will be explained using Figure 6(A). 600 is the setting screen displayed on the display device of the UI device 103. In Embodiment 1, the detection threshold setting for each behavior and the instruction to start the test are controlled using the display screen of the display device of the UI device 103.

[0039] In step S404, clicking 601 in Figure 6(A) selects the imaging device to be used for threshold setting. The options for 601 are a list of imaging devices registered in the image processing system 100.

[0040] In step S405 (acquisition step), clicking 602 or 603 in Figure 6(A) allows the user to select whether the video source used for the test is the live video from the selected imaging device or a recorded video file. This then allows the user to acquire the captured image from the imaging device.

[0041] If 603 is selected, then 604 is clicked to specify the target video file from the folder. In other words, in Embodiment 1, one imaging device can be selected from among multiple imaging devices, and detection thresholds for each behavior can be set for each selected imaging device.

[0042] In step S406 (setting step), the threshold values ​​for the behaviors to be detected are set by moving the sliders 605, 606, 607, and 608 in Figure 6(A) to the left or right, respectively. That is, the detection threshold values ​​for each behavior in the detection step are set. While step S406 is being executed, the UI device 103 functions as a setting means for setting the detection threshold values ​​for each behavior in the behavior detection unit 106.

[0043] Note that the types (number) of behaviors to be detected and the threshold levels for those behaviors are not limited to the example in Figure 6(A). In the example in Figure 6(A), 605 is the slider for setting the threshold for head shaking, 606 is the slider for setting the threshold for behavior A, 607 is the slider for setting the threshold for behavior B, and 608 is the slider for setting the threshold for behavior C.

[0044] These sliders 605-608 function as multiple first-order units for changing the detection thresholds for multiple behaviors. Behaviors A-C may include actions such as taking hands in and out of pockets, turning around, or walking back and forth in the same place, and can be pre-registered as suspicious behavior patterns.

[0045] In step S407, the test is started by clicking one of the test buttons 609 to 612 in Figure 6(A). In the example in Figure 6(A), 609 is the test button corresponding to head shaking, 610 is the test button corresponding to behavior A, 611 is the test button corresponding to behavior B, and 612 is the test button corresponding to behavior C.

[0046] Here, the test buttons 609-612 function as multiple second instruction units, each for performing tests to detect multiple behaviors. The communication unit 109 of the image processing device 101 receives information from the UI device 103 regarding the imaging device to be tested, the behavior of the target, and the threshold value of the target behavior.

[0047] In step S407, the behavior detection unit 106 detects the specified behavior based on this information and transmits the behavior detection result to the UI device 103 via the communication unit 109. If there is insufficient information necessary for behavior detection, this fact is transmitted to the UI device 103 via the communication unit 109. Here, step S407 functions as a detection step that detects a predetermined number of behaviors of the human body based on the movements of the human body obtained from the captured image.

[0048] In step S408, the results of the behavior detection sent from the image processing device 101 are displayed on the test screen. If a communication error occurs at the start of the test or during the test, the process returns to step S406. Note that since the image processing device 101 is in a warning stop state during the test, no warning instruction is sent to the warning device 104 even if behavior detection is performed.

[0049] In step S509 of Figure 5, it is determined whether live video is selected as the video source. In Figure 6(A), live video from the camera is selected, so it is determined to be Yes, and the process proceeds to step S510. In step S510, the live video from the selected imaging device is displayed.

[0050] Figures 7(A) and 7(B) show examples of confirmation screens for behavior detection on the UI device 103. In step S510, the live video from the selected imaging device is displayed on screen 700 of the UI device 103, as shown in screen 701 of Figure 7(A). In other words, the display device of the UI device 103 is capable of displaying the captured image captured by the imaging device.

[0051] Furthermore, as shown in 702, the type of behavior detection currently in progress is displayed on screen 700. Also, as shown in 704, the ID (camera name, etc.) of the target camera is displayed. 705 is the recording start button, and 706 is the button to close the screen.

[0052] Furthermore, if, for example, a head shake test can detect a movement, in Embodiment 1, a highlight frame 703 is displayed around the head of the human body, as shown in Figure 7(A). This indicates on the display unit whether or not the movement detection unit was able to detect the movement. However, the method of displaying whether or not the movement detection was successful is not limited to this. For example, if the movement detection is the act of putting a hand in or out of a pocket, a highlight frame or the like may be displayed on the image around the pocket. If detection is not possible, a warning is displayed.

[0053] During testing, the user can perform actions within the field of view of the target imaging device to confirm whether detection is possible. To avoid the need to perform the same actions each time the threshold is changed, the displayed video can be recorded and saved as a video file on HDD204 after recording is stopped. By playing back the saved demonstration video, it is possible to visually confirm what level of actions can be detected.

[0054] In Figure 7(A), 705 is the button for recording as described above. If the image processing device 101 notifies that there is insufficient information necessary for detecting the behavior, this will be displayed on screen 700. Here, the necessary information refers to the image (video) for a predetermined period of time required to detect the predetermined behavior. In other words, if the time required to detect the predetermined behavior has not elapsed, a warning will be issued indicating insufficient time.

[0055] After the above display is shown in step S510, if a predetermined time has elapsed, step S512 determines whether to terminate the test. If the answer is No, the process returns to step S510. For example, if button 706 is clicked in step S512, the test is determined to be terminated and the process proceeds to step S514.

[0056] On the other hand, if it is determined in step S509 that recorded video has been selected as the video source, the process proceeds to step S511. In step S511, the image acquisition unit 105 reads the target video stored in the HDD 204 and displays the recorded video on the display device as shown in 701 of Figure 7(A), along with whether or not the target behavior can be detected. Note that the recording start button 705 is not displayed at this time. Other displays are the same as in step S510.

[0057] If, in step S513, you click button 706 in Figure 7(A) to end the test, similar to step S512, then you proceed to step S514, close the test screen in Figure 7(A), and return to the display in Figure 6(A).

[0058] As described above, steps S407 to S514 function as test steps to test whether the detection step can be performed for each behavior. In addition, the CPU of the image processing system functions as a test means during this process.

[0059] In step S515, clicking the registration button 613 in Figure 6(A) saves the threshold settings for each behavior of each imaging device shown in Figure 6(A). However, clicking the cancel button 614 will not save the settings.

[0060] When the state of the image processing device 101 is transitioned to the operating state in step S516, behavior detection, suspicious behavior determination, and warning to the warning device 104 are performed based on the behavior detection threshold set in step S515. As described above, according to Embodiment 1, in the suspicious behavior detection system, the threshold for each behavior used in determining suspicious behavior can be set while testing whether each behavior can be detected.

[0061] (Embodiment 2) Embodiment 1 described an example of testing based on behavior. While testing based on behavior is easy for users to understand because the display is simple, it can also be cumbersome because it requires testing for each behavior.

[0062] Furthermore, the ability to detect suspicious behaviors that are combinations of individual behaviors cannot be tested in Embodiment 1. Therefore, in Embodiment 2, the detection status of each behavior is displayed on the UI device 103 at once, and the ability to determine whether suspicious behaviors have occurred is also made available.

[0063] Figure 8 is a flowchart showing an example of operation in the UI device 103 according to Embodiment 2, and Figure 9 is a flowchart continuing from Figure 8. Hereinafter, the detection of each behavior and the method for confirming suspicious behavior in Embodiment 2 will be explained using the flowcharts in Figures 8 and 9 and Figure 6(B) as an example of the operation procedure for the UI device 103. Figure 6(B) is a diagram showing an example of the settings screen in the UI device 103 according to Embodiment 2.

[0064] Furthermore, the CPU, acting as a computer within the image processing system 100, executes the computer program stored in memory, thereby performing the actions shown in each step of the flowcharts in Figures 8 and 9.

[0065] As with Embodiment 1, this embodiment assumes that one or more imaging devices are already registered with the image processing system 100. Furthermore, the processing from steps S801 to S806 is the same as the processing from steps S401 to S406 in Embodiment 1, and therefore its explanation is omitted.

[0066] In step S807 (judgment step), the test for suspicious behavior is started by clicking the test button 615 in Figure 6(B). Here, the test button 615 functions as a third instruction unit to start the test for suspicious behavior. The image processing device 101 receives information about the imaging device under test and information about the threshold values ​​for each behavior set by the sliders 605, 606, 607, and 608 via the communication unit 109.

[0067] Based on this information, the image processing device 101 performs behavior detection and transmits the detection results and whether each behavior is detected or not to the UI device 103 via the communication unit 109. Then, in step S808, the UI device 103 displays a test screen like the one shown in Figure 7(B) based on the received results. Figure 7(B) shows an example of the display of the test results for suspicious behavior detection.

[0068] If there is insufficient information or time to detect each behavior or determine suspicious behavior, the UI device 103 transmits a message to that effect to the UI device 103 via the communication unit 109. In addition, the suspicious behavior determination unit 107 determines suspicious behavior, and if it determines that a person is behaving suspiciously, it transmits that information to the UI device 103 via the communication unit 109.

[0069] Specifically, in step S807, a detection step is performed to detect behavior, and based on the detection results from the detection step, it is determined whether or not there is any suspicious behavior by the human body.

[0070] Similar to Embodiment 1, if a communication error occurs at the start of the test or during the test, the process returns to step S806. Note that since the image processing device 101 is in a warning-stopped state during the test, no warning instructions are sent to the warning device 104 even if behavior detection or suspicious behavior determination is performed.

[0071] Next, in step S909 in Figure 9, it is determined whether live video is selected as the video source. If yes, the process proceeds to step S910. In step S910, the live video from the selected imaging device is displayed on the UI device 103 as shown in 701 of Figure 7(B), and the results of each behavior detection (presence or absence) are displayed as needed.

[0072] As an example of displaying detection results (presence or absence), Figure 7(B) shows that in the 707, the detection results for each behavior are displayed on the right side of the screen, but the display method is not limited to this.

[0073] After the above display is shown in step S910, if a predetermined time has elapsed, step S912 determines whether to terminate the test. If the answer is No, the process returns to step S910. For example, if button 706 is clicked in step S912, the test is determined to be terminated and the process proceeds to step S914.

[0074] Furthermore, if there is an update to the suspicious behavior detection result, the UI device 103 will display the update result, for example, on 707, as needed. In Figure 7(B), a highlight frame is displayed on, for example, the head of a person whose suspicious behavior has been detected, as shown on 703, but the method is not limited to this. During the test, the user will sequentially perform behaviors within the field of view of the selected imaging device, and the detection of each behavior will be confirmed.

[0075] Furthermore, to avoid the need to perform the act multiple times, as in Example 1, recording of the displayed video may be started and saved as a video file to memory such as HDD204 after recording is stopped. Here, memory such as HDD204 functions as a recording means for recording the captured images captured by the imaging device during the test.

[0076] In Figure 7(B), 705 is a button that instructs the start of recording, as described above. If the image processing device 101 notifies that there is insufficient information necessary for detecting behavior or determining suspicious activity, this fact may be displayed on screen 700.

[0077] On the other hand, if recorded video was selected as the video source in step S909, the process proceeds to step S911. In step S911, the image acquisition unit 105 reads the video of the target file stored in the HDD 204 and displays the recorded video on the display device as shown in 701 of Figure 7(B). At this time, the recording start button 705 for instructing the start of recording is not displayed. Other displays are the same as in step S910.

[0078] In step S913, it is determined whether the test is finished, similar to step S912. If the test is ended by clicking, for example, 706 in Figure 7(B), the process proceeds to step S914. In step S914, the screen in Figure 7(B) is closed, and the display returns to Figure 6(B). The processing in steps S914 to S916 is the same as in steps S514 to S516 in Embodiment 1, so the explanation is omitted.

[0079] As described above, according to Embodiment 2, in the suspicious behavior detection system, the detectability of each behavior used to determine suspicious behavior can be checked simultaneously by clicking the suspicious behavior determination test button 615. After checking the results of the suspicious behavior determination, it is possible to set a threshold.

[0080] Furthermore, a storage device may be provided as a means of saving a history of suspicious activity during the period in which the image processing system is operational. For example, an HDD204 may be used as the storage device, or the storage device may be placed on a server connected to a network. However, when performing tests using the test means, it is desirable to control the system so that the history is not saved by the history saving means.

[0081] Although the present invention has been described in detail above based on its preferred embodiments, the present invention is not limited to the above embodiments, and various modifications are possible in accordance with the spirit of the present invention, and these modifications are not excluded from the scope of the present invention. Furthermore, the present invention includes the following combinations.

[0082] (Configuration 1) An image processing system characterized by comprising: acquisition means for acquiring captured images captured by an imaging device; detection means for detecting a predetermined number of behaviors of the human body based on the movements of the human body obtained from the captured images; determination means for determining whether or not there is suspicious behavior of the human body based on the detection results of the detection means; warning means for issuing a warning when the determination means determines that there is suspicious behavior; setting means for setting a detection threshold for each behavior in the detection means; and testing means for testing whether or not the detection means can detect each behavior.

[0083] (Configuration 2) The image processing system according to Configuration 1, characterized in that the setting means has a plurality of first instruction units for changing the detection thresholds of a plurality of behaviors.

[0084] (Configuration 3) The image processing system according to Configuration 1 or 2, characterized in that the setting means has a plurality of second instruction units for each of the plurality of behavior detection tests.

[0085] (Configuration 4) The image processing system according to any one of Configurations 1 to 3, characterized in that the setting means has a third instruction unit for starting a test for determining the suspicious behavior.

[0086] (Configuration 5) The image processing system according to any one of Configurations 1 to 4, characterized in that the acquisition means acquires live video captured by the imaging device or video pre-stored in a storage device as the captured image.

[0087] (Configuration 6) The image processing system according to any one of Configurations 1 to 5, characterized in that the setting means can select one imaging device from among a plurality of imaging devices, and can set a detection threshold for each behavior for each selected imaging device.

[0088] (Configuration 7) When performing the test using the test means, the warning by the warning means is suppressed, as described in any one of Configurations 1 to 6 of the image processing system.

[0089] (Configuration 8) Having at least a history storage means for storing the history of the suspicious behavior, An image processing system according to any one of configurations 1 to 7, characterized in that when the test is performed using the test means, the history is not saved by the history saving means.

[0090] (Configuration 9) An image processing system according to any one of Configurations 1 to 8, characterized in that it has a display means for indicating whether or not the detection by the detection means is possible.

[0091] (Configuration 10) The image processing system according to Configuration 9, characterized in that the display means is capable of displaying the captured image captured by the imaging device.

[0092] (Configuration 11) The image processing system according to Configuration 10, characterized in that the setting means and the test means are controlled using the display screen of the display means.

[0093] (Configuration 12) An image processing system according to any one of Configurations 1 to 11, characterized in that it has recording means for recording the captured images captured by the imaging device during the test.

[0094] (Method) An image processing method characterized by comprising: an acquisition step of acquiring an image captured by an imaging device; a detection step of detecting a predetermined number of behaviors of the human body based on the movements of the human body obtained from the image; a determination step of determining whether or not there is suspicious behavior of the human body based on the detection results of the detection step; a warning step of issuing a warning if it is determined that there is suspicious behavior in the determination step; a setting step of setting a detection threshold for each of the behaviors in the detection step; and a test step of testing whether or not the detection step can detect each of the behaviors.

[0095] A computer program for controlling each of the means of the image processing system described in any one of configurations 1 to 12.

[0096] Furthermore, in order to realize some or all of the control in this embodiment, a computer program that realizes the functions of the above-described embodiment may be supplied to an image processing system, etc., via a network or various storage media. The computer (or CPU, MPU, etc.) in the image processing system, etc., may then read and execute the program. In that case, the program and the storage medium storing the program constitute the present invention. [Explanation of symbols]

[0097] 100: Image Processing System 101: Image Processing Device 102: Imaging device 103:UI device

Claims

1. An acquisition means for acquiring captured images taken by an imaging device, A detection means for detecting a predetermined number of human body movements based on the human body movements obtained from the captured image, A determination means for determining whether or not there is suspicious behavior of the human body based on the detection result by the detection means, If the aforementioned determination means determines that the aforementioned suspicious behavior has occurred, a warning means provides a warning, A setting means for setting a detection threshold for each behavior in the detection means, An image processing system characterized by having a test means for testing whether the detection means can detect each of the aforementioned behaviors.

2. The image processing system according to claim 1, characterized in that the setting means causes a plurality of first instruction units for changing a plurality of detection thresholds for the plurality of behaviors to be displayed on the display unit.

3. The image processing system according to claim 1, characterized in that the setting means displays a plurality of second instruction units on the display unit for each of the plurality of behavior detection tests.

4. The image processing system according to claim 1, characterized in that the setting means causes a third instruction unit for starting a test of determining suspicious behavior to be displayed on the display unit.

5. The image processing system according to claim 1, characterized in that the acquisition means acquires live video captured by the imaging device or video pre-stored in a storage device as the captured image.

6. The image processing system according to claim 1, characterized in that the setting means can select one imaging device from among a plurality of imaging devices, and can set a detection threshold for each behavior for each selected imaging device.

7. The image processing system according to claim 1, characterized in that when the test is performed by the test means, the warning by the warning means is suppressed.

8. The system has at least a history storage means for storing a history of the aforementioned suspicious behavior, The image processing system according to claim 1, characterized in that when the test is performed using the test means, the history is not saved using the history saving means.

9. The image processing system according to claim 1, further comprising a display means for indicating whether or not the detection by the detection means is possible.

10. The image processing system according to claim 9, characterized in that the display means is capable of displaying the captured image captured by the imaging device.

11. The image processing system according to claim 10, characterized in that the setting means and the testing means are controlled using the display screen of the display means.

12. The image processing system according to claim 1, characterized in that it has recording means for recording the captured images captured by the imaging device during the test.

13. The acquisition means includes an acquisition step of acquiring an image captured by an imaging device, The detection means includes a detection step in which it detects a predetermined number of behaviors of the human body based on the movements of the human body obtained from the captured image, The determination means includes a determination step in which it determines whether or not there is suspicious behavior of the human body based on the detection result of the detection step, A warning step in which a warning is issued when the warning means determines that the suspicious behavior has occurred in the determination step, The setting means includes a setting step for setting a detection threshold for each behavior in the detection step, An image processing method characterized in that the test means includes a test step for testing whether the detection step can be performed for each of the behaviors.

14. A computer program for controlling each means of the image processing system according to any one of claims 1 to 12 by a computer.

Images