Imaging apparatus and image analysis method

Abstract

To appropriately detect an event through image analysis even if a specific event occurs while a camera head is not directed to each of a plurality of preset positions, appropriately detect the event through image analysis when detecting the specific event through image analysis at the plurality of preset positions.SOLUTION: An imaging apparatus performs pan-tilt operations to move a camera head over a photographable area, and when the camera head is at a predetermined preset position, stores background image data in a video area at the preset position in a storage device. When the camera head is at the predetermined preset position, the imaging apparatus reads out the background image data in the video area at the preset position stored in the storage device in the past, takes a difference between a value of image data in the video area currently being photographed and a value of the background image data in the video area at the preset position stored in the storage device in the past, and when the difference is equal to or more than a predetermined threshold, determines the presence of an image change.SELECTED DRAWING: Figure 5

Description

[Technical Field]

[0001] The present invention relates to an imaging device and an image analysis method, and more particularly to an imaging device capable of pan-tilt driving that is suitable for use in surveillance applications such as detecting the removal of an article, and an image analysis method. [Background technology]

[0002] In recent years, pan-tilt cameras, which allow users to control the camera and view images via a network, a dedicated line, a remote control, or the like, have become widely used for surveillance purposes. Pan-tilt cameras have a mechanism that allows them to rotate in the pan direction (horizontal) and tilt direction (vertical), and have the function of freely changing the shooting direction and shooting angle of view through user operation or preset operation. Pan-tilt cameras are also known to have a preset position function that allows multiple pan-tilt shooting positions to be registered as preset positions, and when a specific preset position is specified, the shooting position is moved to the registered pan-tilt position. When a specific preset position is set, the pan-tilt camera moves the shooting position to match the pan coordinates and tilt coordinates of the previously registered preset position.

[0003] On the other hand, among the image analysis technologies installed in cameras, there is known an image analysis detection technology that can detect changes in the image when a specific item such as a painting or a bag is removed from a certain area in order to monitor theft, etc.

[0004] When detecting the removal of an item in this way, the pan-tilt camera first creates and saves in advance background image information of the image area before the object is removed, which is the image area showing the object whose removal is to be detected.The saved background image information is then successively compared with the detection area of ​​the current image, and if the difference is equal to or greater than a specific value, a detection event is generated to notify the user that the object has been removed.

[0005] For example, in the conventional technology disclosed in Patent Document 1, when detecting a specific event by image analysis at a preset position, image analysis is performed and the specific event is detected after focus adjustment is completed. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Laid-Open No. 2002-51329 Summary of the Invention [Problem to be solved by the invention]

[0007] However, in general, when the camera is not moved to a predetermined preset position, an object may be taken away. In such a case, the conventional technology disclosed in Patent Document 1 may not be able to detect whether an object has been taken away.

[0008] The object of the present invention is to provide an imaging device that can appropriately detect image changes through image analysis when detecting specific events through image analysis at multiple preset positions, even if an event occurs while the camera has not moved to each preset position. [Means for solving the problem]

[0009] The configuration of the imaging device of the present invention is preferably an imaging device in which the imaging means is capable of moving among a plurality of preset positions, and is equipped with an acquisition means for acquiring an imaged image, a holding means for acquiring and holding a background image at a predetermined preset position when the imaging means is at the predetermined preset position, and a determination means for determining whether an object has been taken away based on difference information between the imaged image at the predetermined preset position and the background image that has been previously held. [Effects of the Invention]

[0010] According to the present invention, an imaging device can be provided that, when detecting specific events through image analysis at multiple preset positions, can appropriately detect image changes through image analysis even if an event occurs while the camera has not moved to each preset position. [Brief explanation of the drawings]

[0011] [Figure 1] 1 is a functional configuration diagram of an imaging device according to a first embodiment and a monitoring system including the imaging device; [Figure 2] FIG. 1 is a diagram illustrating the hardware and software configuration of an imaging device according to a first embodiment. [Figure 3] FIG. 1 is an external view of an imaging device. [Figure 4A] 1A to 1C are diagrams for explaining image change detection processing in the imaging device of the first embodiment (part 1); [Figure 4B] 10A to 10C are diagrams illustrating the image change detection process in the imaging device of the first embodiment (part 2). [Figure 5] 10 is a flowchart illustrating a process of notifying an article being taken away by detecting a change in an image of the imaging device according to the first embodiment. [Figure 6] FIG. 10 is a sequence diagram showing the process of image change detection for each preset position. [Figure 7] FIG. 10 is a functional configuration diagram of an imaging device according to a second embodiment and a monitoring system including the imaging device. [Figure 8] FIG. 10 is a diagram illustrating the hardware and software configuration of an imaging device according to a second embodiment. [Figure 9] 10 is a flowchart illustrating a process of notifying an article being taken away by detecting a change in an image of an imaging device according to a second embodiment. [Figure 10] 10 is a flowchart showing details of background image correction processing. DETAILED DESCRIPTION OF THE INVENTION

[0012] Hereinafter, each embodiment of the present invention will be described with reference to FIGS.

[0013] [Embodiment 1]

[0014] A first embodiment of the present invention will be described below with reference to FIGS. 1 to 5. FIG. First, the configuration of the imaging apparatus of the first embodiment will be described with reference to FIGS. FIG. 1 is a functional configuration diagram of an imaging device according to a first embodiment and a monitoring system including the imaging device. FIG. 2 is a diagram showing the hardware and software configuration of the imaging device of the first embodiment. FIG. 3 is an external view of the imaging device.

[0015] As shown in FIG. 1, the monitoring system of this embodiment has an imaging device 100 and a management terminal 10 connected via a network 5.

[0016] The imaging device 100 of this embodiment is assumed to be a camera capable of pan-tilt operation, which allows the imaging area to be varied by horizontal (pan) and vertical (tilt) movement. The management terminal 10 is a general information processing device such as a PC (personal computer), tablet, or smartphone. The management terminal 10 can transmit, via a network 5, commands for controlling the imaging device 100 and commands for accessing video information captured by the imaging device 100, and can receive video data and image change detection results. The network 5 may be a wired connection or a wireless connection such as Wi-Fi.

[0017] The functional configuration of the imaging device 100 includes a control unit 101, a pan / tilt control unit 110, a pan / tilt driving unit 111, an imaging unit 120, an image processing unit 130, a memory unit 140, an external I / F unit 150, a command analysis unit 151, and a response / data output unit 152.

[0018] The control unit 101 processes each unit of the imaging device 100 and causes them to execute processing. In particular, it issues instructions regarding pan / tilt operations to the pan / tilt control unit 110, executes the image processing unit 130, and performs interface processing with the management terminal 10. The control unit 101 also calculates preset positions. A preset position is a position to which the imaging means of the imaging device 100 is designated to be moved.

[0019] The imaging unit 120 is composed of hardware such as imaging lenses including a focus lens and a zoom lens, an imaging element that converts the optical information collected by the lens into electronic information, and a mechanical drive system and circuits that drive these, and is responsible for capturing images of the subject and converting them into electrical signals.

[0020] The pan / tilt control unit 110 analyzes commands related to pan / tilt control transmitted from the control unit 101, and controls the drive amount, speed, and acceleration / deceleration of the pan / tilt driving unit 111 based on the analyzed command information, thereby performing a pan / tilt operation to move the imaging means. The pan / tilt control unit 110 also performs initialization operations related to the pan / tilt driving unit 111. When a preset position is set, the pan / tilt control unit 110 instructs the pan / tilt driving unit 111 to conform to the preset position.

[0021] The pan / tilt driving unit 111 is realized as hardware by a mechanical driving system that performs pan / tilt operations, a driving motor, a motor driver, etc., and causes the lens to perform pan / tilt operations to move the imaging means.

[0022] The image processing unit 130 generates image data from the electrical signals converted by the imaging unit 120 and performs image processing such as noise removal and gamma correction. In particular, the image processing unit 130 has an image change detection unit 131 as a sub-functional unit. The image change detection unit 131 compares the saved background image data with the currently captured video data at a predetermined timing, such as during initialization or in response to a registration command from the user, and outputs a result indicating that an image change has occurred if there is a certain difference. This makes it possible to detect whether an item has been removed from the imaging area of ​​the imaging device 100.

[0023] The external I / F unit 150 is a functional unit that controls interface processing for exchanging commands and data with external devices such as the management terminal 10 .

[0024] The command analysis unit 151 analyzes commands transmitted from the management terminal 10 via the network 5 and transmits the analyzed commands to the control unit 101. The commands transmitted from the management terminal 10 are, for example, control commands relating to pan / tilt positions or commands for setting preset positions.

[0025] The response / data output unit 152 transmits the response results to commands from the management terminal 10 and the control results of the imaging device 100, and transmits the video data captured by the imaging unit 120 to the management terminal 10 via the network 5.

[0026] The storage unit 140 is a functional unit that stores commands transmitted from the management terminal 10, command processing result data, control data for the imaging device 100, video data captured by the imaging unit 120, and the like. Next, the hardware and software configuration of the imaging device will be described with reference to FIG.

[0027] 2, the imaging device 100 includes, as its hardware configuration, an MPU 201, a main memory device 210, a lens 220, an image sensor 221, an A / D conversion device 222, a network I / F 250, and a non-volatile memory device 260. The imaging device 100 also includes a zoom / focus control mechanism 223, a pan drive mechanism 230, a pan drive motor 231, a tilt drive mechanism 240, and a tilt drive motor 241.

[0028] The MPU (Micro Processor Unit) 201 is a semiconductor memory device that references digital data in the main memory device 210, executes programs consisting of machine instructions, and controls the entire imaging device 100. The main memory device 210 is a volatile semiconductor memory device that stores work data and programs referenced by the MPU 201. The lens group 220 is a set of lenses, usually consisting of multiple lenses, that take in external light and perform enlargement / reduction. The image sensor 221 is a semiconductor sensor that converts light incident from the lens group 220 into an electrical signal. The A / D converter 222 is a circuit that converts analog electrical signals taken in from the image sensor 221 into digital data. The zoom / focus control mechanism 223 is a mechanism that performs zoom processing (enlargement / reduction of imaging) and focus processing (focusing). The pan drive mechanism 230 is a mechanism that is driven by a pan drive motor 231 to align the pan (horizontal direction) of the imaging area. Similarly, the tilt drive mechanism 240 is a mechanism that is driven by a tilt drive motor 241 to align the tilt (horizontal direction) of the imaging area. The network I / F 250 is an interface circuit for exchanging packets with the management terminal 10 via the network 5. The non-volatile storage device 260 is a non-volatile storage device such as a flash memory. Although not shown, the image capture device 100 is equipped with a power supply circuit that converts commercial AC power into DC current and supplies the DC current to each section of the image capture device 100.

[0029] Furthermore, the nonvolatile storage device 260 of this embodiment stores preset position data 400, video data 410, background image data 411, image change detection data 412, received command data 421, and response output data 422 as data.

[0030] The preset position data 400 is data representing a preset position specified by the user. The preset position is defined by the drive position of the tilt drive mechanism 240 and the drive position of the pan drive mechanism 230. The video data 410 is video data captured by the imaging device 100. The background image data 411 is data showing a background image of a certain part of the area that can be captured by the imaging device 100. The image change detection data 412 is data resulting from a comparison between a background image of a certain part of the area that can be captured by the imaging device 100 and an image of that part obtained from the video data 410 at a certain point in time. The received command data 421 is data of a command received from the management terminal. The response output data 422 is response data of the command received from the management terminal.

[0031] The imaging device 100 in this embodiment may have circuits for realizing other functions as hardware. For example, the imaging device 100 may be provided with a video output terminal such as an SDI (Serial Digital Interface) or an HDMI (High-Definition Multimedia Interface) (registered trademark), an audio input / output circuit, or an external device input / output circuit.

[0032] Next, the appearance of the imaging device will be described with reference to FIG.

[0033] The imaging device 100 is a mechanism in which a camera head 505 is mounted on a base 501. The base 501 serves as the base of the device when it is installed on a ceiling or the like, and also functions to support a pan mechanism 502 above (or below) it. The pan mechanism 502 rotates a support column inside the camera head 505 (not shown), causing the camera to pan (swing horizontally) relative to the imaging area.

[0034] Camera head 505 is an imaging means. Lens 504 is installed in camera head 505, and tilt mechanism 503 moves lens 504 to perform tilt (vertical swing) operation relative to the imaging area.

[0035] Next, the image change detection process in the imaging device of the first embodiment will be described with reference to FIGS. 4A and 4B. 4A and 4B are diagrams illustrating the image change detection process in the imaging device of the first embodiment.

[0036] In this embodiment, a situation is assumed in which preset positions A and B are set, and an image 603A that can be captured at preset position A includes an item 600A, and an image 603B that can be captured at preset position B includes an item 600B.

[0037] Imaging device 100 controls pan / tilt driving unit 111 to move camera head 505 to preset position A and preset position B. Imaging device 100 acquires image 603A at preset position A and image 603B at preset position B.

[0038] The imaging device 100 periodically performs a pan-tilt operation within an area where image capture is possible, and moves the camera head 505. The imaging device 100 may also be set to move the camera head 505 to preset position A or preset position B by a pan-tilt operation at a specified time in response to a command from the management terminal 10 or by manual setting.

[0039] Then, the imaging device 100 determines whether the camera head 505 is at preset position A or preset position B, captures an image in the imaging area at the preset position, acquires the captured image, and saves it as background image data. By updating the background image data at the preset position at regular intervals in this way, it is possible to prevent erroneous detection of changes in the image even when the image conditions change gradually, such as when it becomes dusk or when sunlight gradually enters the scene.

[0040] Image 603A that can be captured at preset position A includes item 600A that may be taken away or stolen by a thief or pickpocket. Similarly, image 603B that can be captured at preset position B includes a different item 600B. To determine whether item 600A at preset position A and item 600B at preset position B have been taken away, camera head 505 is moved to preset position A and preset position B by pan-tilt operation. Then, while capturing each image, image change detection unit 131 detects changes in the image to determine whether an item has been taken away. In order for image change detection unit 131 to determine whether an item has been taken away, it is necessary to capture and acquire in advance a background image of the image area including the item to be detected for each preset position.

[0041] An imaging area 601A including article 600A at preset position A is designated in advance by the user as part of the entire image area that can be captured by imaging device 100. Then, by panning and tilting, camera head 505 moves to preset position A and captures background image 602A. If article 600A has not been removed at the time background image 602A is captured, then background image 600A will include article 600A. By continuously determining whether the amount of change in the image obtained from background image 600A and the image of current imaging area 601A exceeds a threshold, image change detection unit 131 can continuously determine whether article 600A has been removed from preset position A.

[0042] This determination of whether or not an item 600B has been removed by detecting a change in image is similarly performed at preset position B. That is, an imaging area 601B including item 600B at preset position B is designated in advance by the user as part of the entire image area that can be captured by imaging device 100. Then, by panning and tilting, camera head 505 moves to preset position B and acquires background image 602B. If item 600B has not been removed at the time background image 602B is acquired, then background image 600B will include item 600B. By continuously determining whether or not the amount of change in the image obtained from background image 600B and the image of imaging area 601B exceeds a threshold, image change detection unit 131 can continuously determine whether or not item 600B has been removed at preset position B.

[0043] For example, suppose that an article 600B is taken away as shown in Fig. 4B. In this case, when an image obtained from the video of image capture area 601B before the article was taken away is compared with a background image of image capture area 601B before the article was taken away, the amount of change is large, and therefore, by appropriately setting a threshold value, it is possible to determine whether the article has been taken away.

[0044] In this embodiment, a background image at each preset position specified by the user is acquired and stored in advance in the non-volatile memory device 260, and the image change detection unit 131 uses this image to compare it with the image obtained from the current video to determine whether or not an item has been taken away from each preset position.

[0045] Next, a process of notifying removal of an article based on image change detection by the imaging device of the first embodiment will be described with reference to FIG. FIG. 5 is a flowchart illustrating a process for notifying an article being taken away by detecting a change in an image of the imaging device according to the first embodiment.

[0046] As shown in FIG. 5, the control unit 101 of the imaging device 100 determines whether or not a command to instruct movement to preset position A or preset position B has been received from the management terminal 10 (S200).

[0047] If a command to move to preset position A or preset position B has not been received (S200: NO), the process ends.

[0048] When a command to move to preset position A or preset position B is received (S200: YES), the control unit 101 of the imaging device 100 determines whether or not image capture is currently being performed at a preset position that is not the target of the movement command (S201). If image capture is not currently being performed at a preset position that is not the target of the movement command (S201: NO), the process proceeds to S203. If image capture is currently being performed at a preset position that is not the target of the movement command (S201: YES), the image captured at the preset position currently being captured is saved as background image data (S202).

[0049] Next, in response to an instruction from the pan / tilt control unit 160, the pan / tilt driving unit 111 performs a pan / tilt operation on the camera head 505, and moves the camera head 505 to the preset position for which a movement instruction command was issued (S203).

[0050] Next, the imaging device 100 reads and acquires background image data at the preset position where the movement instruction command was issued, which is stored in the nonvolatile storage device 260 (S204).

[0051] Next, the imaging device 100 captures an image in the imaging area of ​​the preset position where the pan / tilt operation was performed by the imaging unit 120, and acquires image data (S205). For example, when the movement instruction command is a command to move to preset position B, image data is acquired in imaging area 601B including item 600B at preset position B of video 603B that can be captured at preset position B shown in FIG. 4A.

[0052] Then, the image change detection unit 131 of the image processing unit 130 of the imaging device 100 calculates the difference between the value of the image data of the video captured in S205 and the value of the background image data stored at the preset position acquired in S204, and determines whether the difference is greater than or equal to a threshold value (S206).

[0053] The difference in image data can be evaluated by, for example, taking the absolute value of the difference for each pixel of each image data and summing up the values ​​for the pixels in the target image region, as shown in the following (Equation 1).

number

[0054] Here, PastImPixeli is the pixel value of pixel i in the background image data saved in the past, and PresentImPixeli is the pixel value of pixel i in the image data of the video currently being shot. Also, ThPixel on the right side of (Equation 1) is a threshold, and Σ on the left side indicates that the sum is taken over all the areas being compared.

[0055] If the difference is equal to or greater than the threshold (S206: YES), the management terminal 10 is notified that "an item has been removed" (S207). If the difference is less than the threshold (S206: NO), the management terminal 10 is notified that "an item has not been removed" (S208).

[0056] Next, the imaging device 100 stores the image captured at the preset position in S205 in the non-volatile storage device 260 as background image data (S209).

[0057] Next, the process when the imaging device detects an image change for each preset position will be described with reference to FIG. FIG. 6 is a sequence diagram showing the process of image change detection for each preset position.

[0058] The process described here refers to the process in which the image change detection unit 131 compares the difference between background image data and the image data currently being captured, monitors whether an item to be detected has been removed, and if so, issues an event and notifies the management terminal 10. The imaging device 100 executes processing in units called processes under the operating system (OS), and allocates processor computing resources and memory usage to each process. In an imaging device that performs pan-tilt operations using defined preset positions, as in this embodiment, launching a process for each defined preset position and executing each process independently may reduce the memory usage, computational load, and other resources required for the process.

[0059] In the following, it is assumed that image analysis processes (hereinafter simply referred to as "analysis process A" and "analysis process B") are launched at preset position A and preset position B described in this embodiment, respectively, and that each process operates in response to the preset movement of the camera. Also, a preset position other than preset position A and preset position B is expressed as preset position X.

[0060] When the image change detection unit 131 analyzes image changes at each of the preset positions A and B, the imaging device 100 activates in advance analysis process A and analysis process B as analysis processes for each preset position.

[0061] When the camera head 505 of the imaging device 100 moves to preset position X (S700), analysis process A and analysis process B are both different from their corresponding process set positions, so analysis updates are stopped (S710, S720).

[0062] When the pan / tilt position reaches preset position A due to the pan / tilt operation of the imaging device 100 (S701), analysis process A resumes updating the analysis because it corresponds to the corresponding preset position (S711). Next, when the pan / tilt position starts to move toward preset position B due to the pan / tilt operation of the imaging device 100 (S702), analysis process A stops updating the analysis because the camera head 505 of the imaging device 100 is no longer at its corresponding preset position (S712).

[0063] Thereafter, when the camera head 505 of the image capture device 100 reaches preset position B (S702), analysis process B resumes updating the analysis because the camera head 505 of the image capture device 100 has reached the corresponding preset position (S712). In this way, by resuming the analysis process when a process reaches its corresponding preset position and halting the analysis process when it no longer does so, it is possible to ensure the independence of the processing of each process. Furthermore, each process can retain the internal calculation values ​​used by the image change detection unit 131 for analysis. Therefore, when the update is resumed, the analysis update can be resumed while inheriting the internal calculation values ​​that have already been calculated, which also has the effect of shortening the processing time when the analysis update is resumed.

[0064] As described above, according to this embodiment, when detecting a specific event by image analysis at multiple preset positions where pan-tilt operation is possible, a background image at each preset position is acquired and saved in advance. Therefore, since an image of the image capture area before an item is removed can be acquired, even when a command for removal detection is received from the management terminal at any time, removal of an item from each preset position can be detected.

[0065] [Embodiment 2] A second embodiment of the present invention will be described below with reference to FIGS.

[0066] This embodiment also relates to an imaging device capable of pan-tilt operations similar to those of the first embodiment, and the system configuration is also the same as that of the first embodiment shown in FIG.

[0067] In the first embodiment, the pan / tilt image to be photographed was saved unconditionally, but the imaging device of this embodiment corrects the image when necessary based on data related to the exposure of the imaging device and data on the state of the photographed image. Below, the main differences from the first embodiment will be explained.

[0068] First, the configuration of the imaging apparatus of the second embodiment will be described with reference to FIGS. FIG. 7 is a functional configuration diagram of an imaging device according to the second embodiment and a monitoring system including the imaging device. FIG. 8 is a diagram showing the hardware and software configuration of the imaging device of the second embodiment.

[0069] As shown in FIG. 7, the functional configuration of the imaging device 100 of the second embodiment is such that a background image correction unit 132 is added as a sub-function unit of the image processing unit 130.

[0070] The background image correction unit 132 is a functional unit that corrects the background image used for removal detection at a preset position under specific conditions.

[0071] 8, in the imaging device 100 of this embodiment, exposure data 423 is added as data to be stored in the non-volatile storage device 260 of the first embodiment shown in FIG. 2. The exposure data 423 will be described in detail later.

[0072] Next, a process for notifying removal of an article based on image change detection by the imaging device of the second embodiment will be described with reference to FIGS. FIG. 9 is a flowchart illustrating a process for notifying an article being taken away by detecting a change in an image of the imaging device according to the second embodiment. FIG. 10 is a flowchart showing the details of the background image correction process.

[0073] FIG. 9 is almost the same as the processing in FIG. 5 of the first embodiment, but adds a process of saving the exposure data of the imaging device 100 after saving the background image data, and a process of correcting the background image data based on the exposure data after reading the background image data.

[0074] 9, after S202, a process for saving exposure data of the image capture device (S202s) is added, after S204, a process for correcting background image (S204a) is added, and after S209, a process for saving exposure data of the image capture device (S209s) is added. Furthermore, S206 is changed to S206b. In S206b, the object of comparison is the background image data corrected in S204a.

[0075] The background image correction process (S204a) is a process for correcting the read background image data based on the exposure data and the SN value. Details of the background image correction process will be described later with reference to FIG.

[0076] The process of saving the exposure data of the image capturing device (S202s, S209s) is a process of saving the exposure data of the image capturing device 100 in the non-volatile storage device 260.

[0077] Here, the exposure data is data indicating a numerical value (EV: Exposure Value) that indicates the brightness of the photographing conditions calculated from the shutter speed, gain, aperture, and exposure time of the camera.

[0078] Next, the background image correction process will be described in detail with reference to FIG. This is the process corresponding to S204a in FIG.

[0079] First, the image capturing apparatus 100 reads and acquires the exposure data 423 at the stored preset position (S301).

[0080] Next, the image capturing device 100 reads the saved background image data 411 at the preset position and acquires the SN value (S302). Here, the SN value (also called the S / N ratio) is a numerical value that represents the clarity and sharpness of an image, calculated by taking the ratio of the signal average and deviation in an image information area in which a single white object is captured.

[0081] Next, the imaging device 100 acquires exposure data in the state in which an image is being captured at the current preset position (S303).

[0082] Next, the imaging device 100 calculates the SN value of the image data in the state where the image is being captured at the current preset position (S304).

[0083] Next, image capture device 100 compares the SN values ​​of exposure data 423 read out in S301 and background image data 411 acquired in S302 with the exposure data and the SN values ​​of the image data when capturing images at the current preset position for the same image area (for example, image capture area 601A including item 600A at preset position A in FIG. 4A and image capture area 601B including item 600B at preset position B), and if any of these values ​​is greater than the threshold, determines whether or not it is necessary to correct the background image data (S305). This is because if there is too large a difference between the SN value of background image data previously captured at the preset position, the exposure data when capturing the background image data, the SN value of the image data currently being captured, and the exposure data during current capture, the difference between the image data being compared will be large, regardless of whether the object has been removed, and there is a possibility of false detection.

[0084] The necessity of correcting the background image data in S303 is determined when, for example, the following (Equation 2) and (Equation 3) are satisfied.

number

[0085] Here, PastEV on the left side of (Equation 2) is the value of exposure data (exposure value) saved immediately after background image data was acquired in the past, and PresentEV is the value of exposure data in the current shooting. EV is the threshold value.

[0086] In addition, PastImSN on the left side of (Equation 3) is the SN value of background image data previously stored, and PresentImSN is the SN value of image data currently captured. SN is the threshold value.

[0087] When it is determined based on the above determination conditions that correction of background image data is necessary (S305: YES), the imaging device 100 corrects the background image data (S306).

[0088] At this time, the background image data can be corrected, for example, by the following (Equation 4).

number

[0089] Here, UpdateImPixeli is the pixel value of pixel i of the background image data updated by the correction coefficient, and PastImPixeli is the pixel value of pixel i of the background image data previously saved.

[0090] Furthermore, ΔEV is the left side of the inequality in (Equation 2), and is the difference between the value of the exposure data saved immediately after background image data was acquired in the past and the value of the exposure data in the current shoot, as expressed in (Equation 5) below. ΔImSN is the left side of the inequality in (Equation 3), and is the sum of all the differences between the SN values ​​of the background image data saved in the past and the SN values ​​of the image data in the current shoot, as expressed in (Equation 6), over a certain area being compared.

number

[0091] Furthermore, F(ΔEV, ΔImSN) is a function that determines a correction coefficient with parameters ΔEV and ΔImSN. This function for determining the correction coefficient can be found, for example, from a table that holds values ​​corresponding to pairs of ΔEV and ΔImSN.

[0092] Furthermore, a function for determining the correction coefficient may be explicitly defined by calculating a statistically corresponding formula through big data analysis.

[0093] As described above, according to the image processing of the imaging device of this embodiment, when determining whether an item has been taken away by comparing a background image of a preset position that was saved in the past with an image obtained from the video data currently being captured, the background image of the preset position is corrected based on the difference value between the exposure data and the SN value of the image, so that the determination of whether an item has been taken away can be made appropriately. [Explanation of symbols]

[0094] 5...network, 10...management terminal, 100...imaging device, 101...control unit, 110...pan-tilt control unit, 111...pan-tilt driving unit, 120...imaging unit, 130...image processing unit, 140...storage unit, 150...external I / F unit, 151...command analysis unit, 152...response / data output unit

Claims

1. An imaging device in which an imaging means is movable among a plurality of preset positions, an acquisition means for acquiring a captured image; a holding means for acquiring and holding a background image at a predetermined preset position when the imaging means is at the predetermined preset position; and a determination unit that determines whether an object has been taken away based on difference information between the captured image at the predetermined preset position and the background image that is stored in advance.

2. 2. The imaging device according to claim 1, wherein the determining means determines that the object has been taken away when a difference between the captured image and the background image is equal to or greater than a threshold value.

3. 2. The imaging device according to claim 1, wherein when said imaging means is at a predetermined preset position, said background image at said preset position is periodically saved in a storage device.

4. 2. The imaging device according to claim 1, wherein when the imaging means is at a first preset position and the imaging means moves to a second preset position, background image data in the image area of ​​the first preset position is stored in a storage device.

5. 2. The imaging device according to claim 1, wherein, for each predetermined preset position, an image analysis process corresponding to the respective preset position is performed, and when the imaging means moves from the respective preset position, the image analysis process corresponding to the preset position is stopped.

6. 2. The imaging device according to claim 1, further comprising a correction unit that corrects the background image based on an exposure value.

7. 1. An image analysis method for an imaging device in which an imaging means is movable among a plurality of preset positions, comprising: an acquisition step of the imaging device acquiring an image; a step of the imaging device acquiring and holding a background image at a predetermined preset position when the imaging means is at the predetermined preset position; an image analysis method characterized by comprising a step in which the imaging device determines whether an object has been taken away based on difference information between the captured image at the predetermined preset position and the background image that has been previously stored.

8. 7. A computer program for controlling each unit of the imaging apparatus according to claim 1 by a computer.

Images