Setting method, setting apparatus, and computer program
The method and device for setting image ranges across multiple cameras address the challenges of generating accurate flow lines by adjusting image ranges based on camera configurations and target conditions, resulting in improved flow line accuracy and continuity.
Patent Information
- Application Number
- PCT/JP2024/018133
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2023-06-29
- Filing Date
- 2024-05-16
- Publication Date
- 2025-01-02
AI Technical Summary
Existing methods for generating flow lines using images from multiple cameras face challenges in accurately setting the range of images due to factors like target size, movement speed, environment, and camera installation, leading to narrow or overlapping flow lines.
A method and device that connect multiple cameras to set and adjust the range of images captured by each camera, allowing for the generation of accurate flow lines by detecting movement lines from each camera's frame and adjusting the interval between image ranges to ensure proper combination.
Enables the generation of accurate and continuous flow lines by adjusting the image ranges captured by multiple cameras, addressing issues of narrow or overlapping flow lines and improving the accuracy of target tracking in surveillance systems.
Smart Images

Figure JP2024018133_02012025_PF_FP_ABST
Abstract
Description
Setting method, setting device and computer program
[0001] The present disclosure relates to a setting method, a setting device, and a computer program that use images captured by multiple cameras to set a range of images to be used for generating a target's traffic line.
[0002] There is known a monitoring device that monitors an object moving in a specific space (see, for example, Patent Document 1). For a monitoring camera, it is important to accurately grasp the space of the object. Patent Document 1 discloses correspondence between a map and a camera image.
[0003] Furthermore, there is also known a technique for extracting the movement line of an object moving through a specific space (see, for example, Non-Patent Documents 1 and 2).
[0004] JP 2015-32069 A
[0005] "On-site Cyber-Physical Systemization Technology for B2B," Hidehiko Kanade and 3 others, Panasonic Techniques, Vol. 65 No. 1, May 2019. "Multi-Camera People Tracking with a Probabilistic Occupancy Map," Francois Fleuret et al., March 2007.
[0006] When extracting traffic lines from only one camera, the space in which traffic lines can be generated is limited. Furthermore, it is not easy to combine images captured by multiple cameras. Specifically, the range of images captured by multiple cameras is affected by various conditions, such as the size of the target, the speed at which the target moves, the environment in which the target moves, the camera installation position, and the camera performance.
[0007] The present disclosure provides a setting method, a setting device, and a computer program for setting a range of images to be used for generating a target traffic line using images captured by multiple cameras.
[0008] The setting method disclosed herein is executed in a setting device connected to multiple cameras capturing different areas of a single space, and sets a range of the image to be used for generating a traffic line of an object using images obtained from the multiple cameras. The setting method includes the steps of: acquiring, by the setting device, an image including multiple frames captured by the multiple cameras; extracting a first range from each frame of a first image captured by a first camera; extracting a second range different from the first range from each frame of a second image captured by a second camera capturing an area adjacent to the first camera; detecting a first traffic line of the object from the first range for each frame of the first image; detecting a second traffic line of the object from the second range for each frame of the second image; and resetting the interval between the first range and the second range when the first traffic line and the second traffic line cannot be combined into a single traffic line.
[0009] Such general and specific aspects may be realized by a system, a method, and a computer program, as well as a combination thereof.
[0010] The setting method, setting device, and computer program disclosed herein can use images captured by multiple cameras to set the range of images to be used for generating a target's traffic line.
[0011] 5A shows an example of a first flow line and a second flow line obtained from video captured by two cameras. FIG. 1B shows an example of combining the first flow line and the second flow line shown in FIG. 1A. FIG. 1C shows an example of overlapping of the first flow line and the second flow line. FIG. 1D shows an example of cutting out a partial range from a frame obtained from video captured by two cameras. FIG. 1E is a schematic diagram illustrating video captured by two cameras and a range to be cut out from the video. FIG. 1F is a block diagram showing the configuration of a setting device according to a first embodiment. FIG. 1G is a schematic diagram illustrating an example of area data used by the setting device. FIG. 1H is a schematic diagram illustrating an example of range data used by the setting device. FIG. 1I is a schematic diagram illustrating ranges set on two different cameras. FIG. 5A is a schematic diagram illustrating an example of changing the range of FIG. 5A. FIG. 5I is a schematic diagram illustrating an example of changing the range of FIG. 5A. FIG. 5I is a schematic diagram illustrating an example of changing the range of FIG. 5A. FIG. 5I is a schematic diagram illustrating an example of an order in which ranges are set by the setting device. FIG. 1I is a schematic diagram illustrating another example of an order in which ranges are set by the setting device. FIG. 1I is a flowchart illustrating an example of processing executed by the setting device.
[0012] Hereinafter, embodiments of the present disclosure will be described with reference to the accompanying drawings. However, unnecessary detailed description may be omitted. For example, detailed description of well-known matters or redundant description of substantially identical configurations may be omitted. This is to avoid unnecessary redundancy in the following description and to facilitate understanding by those skilled in the art. Note that the inventors provide the accompanying drawings and the following description to enable those skilled in the art to fully understand the present disclosure, and the subject matter recited in the claims is not limited by them.
[0013] A generation system according to the present disclosure generates a traffic line of a single object from video captured by multiple cameras. In the present disclosure, an "object" refers to a moving object moving within a space. For example, the object may be a human, a robot, a vehicle, or the like. Below, an overview of the processing performed by the generation system will be first described. A specific configuration of the generation system will be described later with reference to FIG. 3. The generation system first generates multiple traffic lines for each video captured by each camera. The generation system then combines the generated multiple traffic lines to generate a single continuous traffic line. The generation system can extract and use specified ranges from frames of video captured by each camera. This allows the generation system to generate an accurate traffic line. A setting method, setting device, and computer program according to the present disclosure set a range to be extracted from each frame of video captured by multiple cameras.
[0014] First, an overview of the process of detecting a single traffic line from the video of a first camera and the video of a second camera will be described. FIG. 1A shows an example of a first traffic line and a second traffic line obtained from video captured by two cameras. As shown in FIG. 1A, the first camera captures a first area F1. The second camera captures a second area F2 adjacent to the first area F1. FIG. 1A shows an example in which an object moves from the first area F1 to the second area F2. The generation system detects points P11 to P15 representing the position of the object from each frame of the video captured by the first camera. The generation system connects the detected points P11 to P15 to generate a first traffic line L1. The generation device also detects points P21 to P25 representing the position of the object from each frame of the video captured by the first camera. The generation device connects the detected points P21 to P25 to generate a second traffic line L2. 1B shows an example of combining the first and second traffic lines shown in FIG. 1A. The generation system combines the first traffic line L1 and the second traffic line L2 by combining points P15 and P21, as shown by the dashed line in FIG. 1B, to generate a single traffic line.
[0015] As shown in FIG. 1A, the areas F1 and F2 captured by each camera overlap. However, it is preferable for the generation system to set the ranges used in generating the traffic lines so that they do not overlap. FIG. 1C shows an example in which the first traffic line and the second traffic line overlap. For example, assume that an object is detected at points P16 and P20 between points P15 and P21 as shown in FIG. 1C. In FIG. 1C, points P16 and P20 are offset for clarity, but in reality, they are at the same location. When images are captured separately by the first and second cameras as shown in FIG. 1C, the generation system detects the objects separately. Therefore, the first traffic line L1 and the second traffic line L2 can be considered separate traffic lines.
[0016] FIG. 1D shows an example of extracting a portion of a frame from video captured by two cameras. As shown in FIG. 1D, ranges R1 and R2 (dashed lines) used for generating a traffic line are set within the areas F1 and F2 (dashed lines) captured by each camera. By appropriately setting the distance D between the first range R1 and the second range R2, the traffic line of the object can be accurately generated. For example, if the first range R1 and the second range R2 overlap, the traffic lines may overlap, as described above in FIG. 1C. On the other hand, if the first range R1 and the second range R2 are too wide, the first traffic line L1 and the second traffic line L2 cannot be connected.
[0017] FIG. 2 shows an example in which an object T is photographed by two cameras 2A and 2B installed on the ceiling. In the example shown in FIG. 2, each of the cameras 2A and 2B is an omnidirectional camera. In FIG. 2, the angle of view of each camera 2A and 2B is indicated by a dashed line. The cameras 2A and 2B photograph a person, the object T, moving in the direction indicated by the dashed arrow. In the example shown in FIG. 2, while the object T is moving, it is captured by one of the cameras 2A and 2B near the waist height H2 of the object T. Therefore, if the object is detected at the waist height H2 of the object T and a movement line is generated, an accurate movement line is likely to be generated. On the other hand, at the height H1 of the object T's feet, the two cameras 2A and 2B overlap for a longer period of time compared to the height H2. Furthermore, at the height H3 of the object's head, the two cameras 2A and 2B do not photograph for a longer period of time compared to the height H2. Therefore, if the object is detected at the height H1 of the object T's feet, there is a risk that a movement line will be generated that overlaps. Furthermore, at height H1 of the feet of target T, it may be difficult to track the target due to surrounding objects, etc. Furthermore, at height H3 of target T's head, there is a risk that the movement line will be interrupted due to a long period of time when neither camera 2A nor 2B is capturing the target. The setting method, setting device, and computer program disclosed herein use images captured by each camera to set a range in which it is easy to track the target, corresponding to the conditions of the target and the environment, etc., as described above.
[0018] [Embodiment] FIG. 3 shows a generation system 100 including a setting device 1 according to an embodiment. The setting device 1 is connected to multiple cameras 2A to 2C via a network 3. The setting device 1 sets a range to be used for generating a target's traffic line for each of the images captured by the cameras 2A to 2C. The setting device 1 also cuts out the set range from the images captured by the cameras 2A to 2C to generate a target's traffic line. FIG. 3 shows an example in which the generation system 100 includes three cameras 2A to 2C. However, the number of cameras connected to the setting device 1 in the generation system 100 is not limited. In the following description, unless there is a particular need to distinguish between them, the cameras will be simply referred to as camera 2.
[0019] The setting device 1 is an information processing device including, for example, an arithmetic circuit 11, an input device 12, an output device 13, a communication circuit 14, and a storage device 15, as shown in FIG.
[0020] The arithmetic circuit 11 is a controller that controls the entire setting device 1. For example, the arithmetic circuit 11 reads and executes a setting program P stored in the storage device 15. This allows the setting device 1 to perform various processes related to setting the range used to generate a traffic line for each of the images captured by each camera. The arithmetic circuit 11 may be various processors such as a CPU, an MPU, a GPU, an FPGA, a DSP, an ASIC, or a dedicated hardware circuit.
[0021] The input device 12 may be an operation button, a keyboard, a touch panel, a microphone, etc., which are used for inputting operations and data. The output device 13 may be a display, a speaker, a lamp, etc., which are used for outputting processing results and data.
[0022] The communication circuit 14 performs data communication between the setting device 1 and external devices such as the cameras 2. The data communication may be wired and / or wireless, and may be in accordance with, for example, a known communication standard. For example, wired data communication may be performed by using a semiconductor integrated circuit communication controller that operates in accordance with the Ethernet (registered trademark) standard and / or the USB (registered trademark) standard as the communication circuit 14. Wireless data communication may be performed by using a semiconductor integrated circuit communication controller that operates in accordance with the IEEE 802.11 standard for wireless local area networks (LANs) and / or the fourth-, fifth-, and sixth-generation mobile communication systems, known as 4G, 5G, and 6G, for mobile communications.
[0023] The storage device 15 is a recording medium for recording various information. The storage device 15 is realized by, for example, RAM, ROM, flash memory, SSD (Solid State Drive), hard disk drive, other storage devices, or an appropriate combination thereof. The storage device 15 stores various data obtained by processes such as generating flow lines and setting ranges in the setting device 1. The storage device 15 also stores the setting program P.
[0024] The area data 151 is data indicating the area of space that each camera 2 captures in the space. FIG. 4A is a schematic diagram showing an example of the area data 151. For example, as shown in FIG. 4A, the area data 151 is data indicating the association between each camera and the area captured by each camera in the space S. FIG. 4A shows the space S as a two-dimensional map. FIG. 4A also shows, with dashed lines, an area F1 captured by the first camera 2A and an area F3 captured by the third camera 2C in the two-dimensional space S. FIG. 4A also shows, with dashed lines, an area F2 captured by the second camera in the two-dimensional space S. For example, the area data 151 is generated in accordance with an operator's operation.
[0025] The range data 152 indicates the range of the area captured by each camera to be used for generating a traffic line. FIG. 4B is a schematic diagram illustrating an example of range data. For example, as shown in FIG. 4B, the range data 152 indicates the association between each camera in a space S and the ranges cut out from the images captured by each camera for generating a traffic line. FIG. 4B illustrates, in a two-dimensional representation of the space S, a range R1 cut out from the frame captured by the first camera 2A and a range R3 cut out from the frame captured by the third camera, indicated by dashed lines. FIG. 4A also illustrates, in a two-dimensional representation of the space S, a range R2 cut out from the frame captured by the second camera 2B, indicated by a dashed-dotted line. For example, the range data 152 is generated in the setting device 1 based on the area data 151. By using the range set by the range data 152, it is possible to appropriately generate a traffic line of the target.
[0026] The following describes each process for setting the range of the image to be used for generating a traffic line, which is executed by the arithmetic circuit 11. The arithmetic circuit 11 also generates a traffic line in the following process. However, the traffic line generated in the following process is generated in order to set the range to be used for generating a traffic line. The generation of a traffic line for actual traffic line analysis, etc., is executed after the range setting described below.
[0027] (Initial Setting) The arithmetic circuit 11 receives an initial setting signal for each camera 2 via the input device 12 through an operation by an operator. The arithmetic circuit 11 sets parameters for each camera using the received setting signal. For example, the camera parameters include the camera position, focal length, etc. Alternatively, a perspective projection matrix may be used as the camera parameters. The arithmetic circuit 11 also generates area data 151 using the received setting signal. Specifically, the arithmetic circuit 11 associates the area photographed by each camera 2 with a map that represents a two-dimensional space based on the setting signal. The arithmetic circuit 11 also stores the generated area data 151 in the storage device 15. After the initial setting, each camera 2 starts capturing video.
[0028] (Acquisition of Images) The arithmetic circuit 11 acquires images captured by each camera 2 via the communication circuit 14. The arithmetic circuit 11 stores the acquired images in the storage device 15. The images acquired by the arithmetic circuit 11 here are images of an object moving in space. Note that the arithmetic circuit 11 may acquire images in advance and read and use past images stored in the storage device 15.
[0029] (Selection of Two Cameras) The arithmetic circuit 11 refers to the area data 151 and the range data 152 and selects two cameras 2. Here, of the two cameras 2 capturing images of adjacent areas, the arithmetic circuit 11 selects two cameras 2 for which ranges to be used for generating traffic lines have not been set. Assume that the area data 151 includes areas F1 to F3 shown in FIG. 4A, and that ranges to be used for generating traffic lines have not been set for any of the cameras 2A to 2C. In this case, the arithmetic circuit 11 selects, for example, the first camera 2A and the second camera 2B capturing images of adjacent areas.
[0030] (Accepting Time) The arithmetic circuit 11 accepts the time required to move from the shooting range of the first camera to the shooting range of the second camera of the two selected cameras. The time is input by an operator via the input device 12. The arithmetic circuit 11 uses the accepted travel time to generate a subsequent flow line. Specifically, the accepted travel time is used to determine whether the flow lines are connected into one line within the accepted travel time.
[0031] Here, the shooting range of the first camera and the shooting range of the second camera have not been determined at the timing when the arithmetic circuit 11 receives the movement time. Therefore, for example, the arithmetic circuit 11 uses the provisional shooting range of the first camera as the shooting range of the first camera and the provisional shooting range of the second camera as the shooting range of the second camera. The provisional shooting range of the first camera and the provisional shooting range of the second camera may be specified, for example, by an operator's operation. Alternatively, the provisional shooting range of the first camera and the provisional shooting range of the second camera may be the shooting area of the first camera and the shooting area of the second camera indicated by the area data 151.
[0032] 4A , the shooting area of the first camera and the shooting area of the second camera indicated by the area data 151 may overlap. Therefore, when each area indicated by the area data 151 is set as a provisional shooting range of the first camera and a provisional shooting range of the second camera, the arithmetic circuit 11 may accept values such as a shooting time range or a video frame number as the time required to move from the shooting range of the first camera to the shooting range of the second camera. When there is a time when an object is photographed only by the first camera and a time when an object is photographed only by the second camera, each area indicated by the area data 151 can be set as a provisional shooting range of the first camera and a provisional shooting range of the second camera.
[0033] (Detecting an Object) The arithmetic circuit 11 uses a predetermined algorithm to detect an object from each frame of video captured by each camera 2. At this time, the arithmetic circuit 11 detects the coordinates of the object from a range cut out from each frame. Specifically, the arithmetic circuit 11 cuts out a predetermined rectangular range from the entirety of each frame of video as a provisional range to be used for generating a traffic line. The arithmetic circuit 11 may also detect an object from the provisional range cut out from each frame. The predetermined range may be the entire area captured by each camera, or at least a portion of the area. For example, a portion of the entire area captured by each camera 2, reduced by a predetermined percentage, may be set as the provisional range.
[0034] For example, as shown in FIG. 4A , the arithmetic circuit 11 cuts out a rectangular portion of a first predetermined size from region F1 of each frame of the first video captured by the first camera 2A as a provisional first range. The first predetermined size may be the same size as region F1 of the first camera's video, or may include at least a portion of region F1. Furthermore, the arithmetic circuit 11 cuts out at least a portion of a rectangular portion of a second predetermined size, different from the first provisional range, from region F2 of each frame of the second video captured by the second camera 2B, which captures an area adjacent to the first camera 2A, as a provisional second range. The second predetermined size may be the same size as region F2 of the second camera's video, or may include at least a portion of region F2. FIG. 5A shows an example of a provisional first range R1′ and a provisional second range R2′. When the tentative first range R1' and the tentative second range R2' are determined as the actual first range R1 and the actual second range R2 in the subsequent processing, they are stored in the storage device 15 as range data 152.
[0035] (Generating Traffic Lines) The arithmetic circuit 11 generates traffic lines of objects for the images captured by each camera 2. Specifically, the arithmetic circuit 11 generates a first traffic line by combining the coordinates of objects detected from the tentative first range in each frame of the first camera 2A. The arithmetic circuit 11 also generates a second traffic line by combining the coordinates of objects detected from the tentative second range of the second camera 2B. In generating the traffic lines, the arithmetic circuit 11 uses the "travel time" for moving from the capture range of the first camera to the capture range of the second camera, which was received in the above-mentioned process. Using the travel time, it is possible to determine whether the first traffic line and the second traffic line are connected into one line.
[0036] (Determination) The arithmetic circuit 11 determines whether the traffic lines generated from the images of the two cameras 2 are combined, overlap, or are disconnected. When the first traffic line and the second traffic line are correctly combined, the arithmetic circuit 11 generates range data 152 in which the tentative first range R1′ is the first range R1 and the tentative second range R2′ is the second range R2, and stores the range data 152 in the storage device 15.
[0037] When the first and second flow lines are disconnected and not connected, the arithmetic circuit 11 narrows the distance D between the provisional first range R1' and the provisional second range R2'. For example, the arithmetic circuit 11 may change the state shown in FIG. 5A to that shown in FIG. 5B. Specifically, the new provisional first range R1' may be set by enlarging the size of the side of the provisional first range R1' that is close to the provisional second range R2' so that it approaches the provisional second range R2'. In this case, the arithmetic circuit 11 may increase the size of at least one of the provisional first range R1' and the provisional second range R2' by a predetermined step size. By enlarging the provisional first range R1', the new distance D' between the provisional first range R1' and the provisional second range R2' becomes narrower than the distance D. Furthermore, for example, the arithmetic circuit 11 may set the new provisional second range R2' by sliding the provisional second range R2' closer to the provisional first range R1', as shown in FIG. 5C. At this time, the arithmetic circuit 11 may slide at least one of the tentative first range R1′ and the tentative second range R2′ toward the other in a predetermined step width. By sliding the tentative second range R2′, a new distance D′ between the tentative first range R1′ and the tentative second range R2′ becomes smaller than the distance D.
[0038] When the first flow line and the second flow line overlap, the arithmetic circuit 11 sets the distance D between the provisional first range R1' and the provisional second range R2' to be larger. For example, as shown in FIG. 5D , the arithmetic circuit 11 may set a new provisional first range R1' by reducing the size of the side of the provisional first range R1' that is close to the provisional second range R2' so that it is farther away from the provisional second range R2'. In this case, the arithmetic circuit 11 may narrow the size of at least one of the provisional first range R1' and the provisional second range R2' by a predetermined step width. By reducing the provisional first range R1', the new distance D' between the provisional first range R1' and the provisional second range R2' becomes wider than the distance D. Also, for example, as shown in FIG. 5E , the arithmetic circuit 11 may set a new provisional second range R2' by sliding the provisional second range R2' away from the provisional first range R1'. At this time, the arithmetic circuit 11 may slide at least one of the tentative first range R1′ and the tentative second range R2′ in a predetermined step width so as to move it away from the other. By sliding the tentative second range R2′, a new distance D′ between the tentative first range R1′ and the tentative second range R2′ becomes wider than the distance D.
[0039] When the tentative first range R1' and the tentative second range R2' are changed, the calculation circuit 11 generates new first and second movement lines. The calculation circuit 11 also determines whether the newly generated first and second movement lines are correctly combined. The calculation circuit 11 continues updating the tentative first range R1' and the tentative second range R2' until the first and second movement lines are correctly combined. When the first and second movement lines are correctly combined, the calculation circuit 11 determines the tentative first range R1' as the first range R1 and the tentative second range R2' as the second range R.
[0040] After setting the new tentative first range R1' and the new tentative second range R2', the arithmetic circuit 11 again generates the first flow line L1 and the second flow line L2. The arithmetic circuit 11 also determines whether the newly generated first flow line L1 and second flow line L2 are correctly connected.
[0041] Note that FIG. 4B illustrates an example in which one side of the first range R1 is adjacent to only one side of the second range R2. However, as shown in FIGS. 6A and 6B, one side of one range may be adjacent to the sides of multiple ranges. FIGS. 6A and 6B are schematic diagrams illustrating an example of the order in which the setting device sets the ranges. In such an example, when changing the provisional range, it is necessary to change it while minimizing the impact on other areas. For example, in the example of FIG. 6A, assume that provisional shooting range A and provisional shooting range B are selected as targets for change. In this case, changing provisional shooting range B moves the position of side sb, which affects another provisional shooting range C. Therefore, provisional shooting range A is changed.
[0042] Also, for example, in the example of FIG. 6B , assume that in the first stage, virtual shooting range A and virtual shooting range D are selected as targets for change. At this time, virtual shooting range D is changed to match the position of side sa without changing virtual shooting range A. In the second stage, virtual shooting range A and virtual shooting range B are selected, and virtual shooting range B is changed to match the position of side sa without changing virtual shooting range A. In the third stage, virtual shooting range B and virtual shooting range C are selected, and virtual shooting range C is changed to match the position of side sb without changing virtual shooting range B. In the fourth stage, virtual shooting range C and virtual shooting range E are selected, and virtual shooting range E is changed to match the position of side sc without changing virtual shooting range C. Once the position of a fixed side is changed, other fixed sides are affected. Therefore, other ranges are changed without changing the position of a fixed side. Note that in the examples of FIG. 6A and FIG. 6B , only sides facing each other vertically in the figures have been considered, but sides facing each other horizontally are also adjusted in the same way. For example, the distance between the provisional photographing range A and the provisional photographing range C is adjusted in the same manner.
[0043] <Range Setting Process> The flow of the process in the setting device 1 will be described using the flowchart shown in Fig. 7. The specific process of each step in the flowchart has been described above, so each step will be described in a simplified manner.
[0044] The arithmetic circuit 11 sets camera parameters (S01).
[0045] The arithmetic circuit 11 also sets the initial position of the camera 2 (S02), thereby generating area data 151 and storing it in the storage device 15.
[0046] The arithmetic circuit 11 acquires images from each camera 2 (S03).
[0047] The arithmetic circuit 11 selects two nearby cameras 2 for which no range is set for use in generating a flow line (S04).
[0048] The arithmetic circuit 11 receives the time required to move from the provisional range of the first camera to the provisional range of the second camera of the two cameras selected in step S04 (S05).
[0049] The arithmetic circuit 11 detects the coordinates of the target from the provisional range of each frame of the video (S06).
[0050] The arithmetic circuit 11 combines the coordinates of each frame obtained in step S06 to generate a first flow line and a second flow line, and also combines the first flow line and the second flow line (S07).
[0051] The arithmetic circuit 11 determines a flow line in which the first flow line and the second flow line are combined in step S07 (S08).
[0052] If the result of the determination in step S08 is that the first and second flow lines are correctly connected without any interruptions or overlaps, the arithmetic circuit 11 determines the tentative first range and the tentative second range as the first range and the second range (S09). At this time, the arithmetic circuit 11 generates range data 152 including the first range and the second range, and stores it in the storage device 15.
[0053] After the first and second ranges are determined, it is determined whether the setting of the ranges to be used for generating the flow lines of all cameras has been completed (S10). If not completed (NO in S10), the processing returns to step S04, and the arithmetic circuit 11 executes processing for the ranges that have not been determined.
[0054] If the result of the judgment in step S08 is that there is a break in the first and second movement lines, the calculation circuit 11 widens the interval between the provisional first range and the provisional second range and sets new provisional first range and provisional second range (S11).
[0055] If the result of the judgment in step S08 is that there is an overlap between the first and second movement lines, the calculation circuit 11 narrows the interval between the tentative first range and the tentative second range and sets new tentative first range and tentative second range (S12).
[0056] When the setting of the ranges to be used for generating the flow lines of all cameras has been completed (YES in S10), the process ends.
[0057] As described above, the setting device 1 according to the embodiment uses images captured by multiple cameras to set the range of images to be used for generating the traffic line of the target. As a result, the generation system 100 can accurately generate the traffic line by using the set range.
[0058] Overview of the Embodiments (1) A setting method disclosed herein is a setting method executed in a setting device that is connected to multiple cameras that capture different areas of a single space, and that uses images obtained from the multiple cameras to set a range of images to be used for generating a movement line of an object, the setting device: acquiring an image including multiple frames captured by the multiple cameras; extracting a first range from each frame of a first image captured by a first camera; extracting a second range different from the first range from each frame of a second image captured by a second camera that captures an area adjacent to the first camera; detecting a first movement line of the object from the first range for each frame of the first image; detecting a second movement line of the object from the second range for each frame of the second image; and resetting the interval between the first range and the second range when the first movement line and the second movement line cannot be combined into a single movement line.
[0059] This allows the range used to generate the target flow line to be appropriately set.
[0060] (2) The generation method of (1) may be configured such that, when the first flow line and the second flow line cannot be combined into a single flow line, the setting device narrows the distance between the first range and the second range.
[0061] This allows the range used to generate the target flow line to be appropriately set.
[0062] (3) In the generation method of (1) or (2), the setting device may be configured to widen the distance between the first range and the second range when the first traffic line and the second traffic line overlap.
[0063] This allows the range used to generate the target flow line to be appropriately set.
[0064] (4) In the generation methods of (1) to (3), when the first flow line and the second flow line are combined, the setting device newly determines the first range determined in the previous processing as the first range, and sets the second range determined in the previous processing as the second range.
[0065] This allows the range used to generate the target flow line to be appropriately set.
[0066] (5) In the generation methods of (1) to (4), when the setting device sets the interval between the first range and the second range to be narrow, the size of at least one of the first range and the second range may be increased.
[0067] This allows the range used to generate the target flow line to be appropriately set.
[0068] (6) In the generation methods of (1) to (5), when the setting device sets the interval between the first range and the second range to be wider, the size of at least one of the first range and the second range may be reduced.
[0069] This allows the range used to generate the target flow line to be appropriately set.
[0070] (7) In the generation methods of (1) to (6), when the setting device is used to set the interval between the first range and the second range to be narrower, at least one of the first range and the second range may be slid closer to the other range or so as to overlap partially.
[0071] This allows the range used to generate the target flow line to be appropriately set.
[0072] (8) In the generation methods of (1) to (7), when the distance between the first range and the second range is widened, the setting device may be set to slide at least one of the first range and the second range so as to approach the other range.
[0073] This allows the range used to generate the target flow line to be appropriately set.
[0074] (9) In the generating methods of (1) to (8), the subject may be a person.
[0075] This allows the range used to generate the flow lines of people to be set appropriately.
[0076] (10) In the generating methods of (1) to (9), the image may be an image obtained by an omnidirectional camera.
[0077] This makes it possible to appropriately set the range to be used for generating the movement line of the object using the image obtained by the omnidirectional camera.
[0078] (11) A setting device disclosed herein is connected to a plurality of cameras that capture different areas of a single space, and uses images obtained from the plurality of cameras to set a range of the image to be used for generating a movement line of an object, the setting device comprising an arithmetic circuit, which acquires an image including a plurality of frames captured by the plurality of cameras, cuts out a first range from each frame of a first image captured by a first camera, cuts out a second range different from the first range from each frame of a second image captured by a second camera that captures an area adjacent to the first camera, detects a first movement line of the object from the first range for each frame of the first image, detects a second movement line of the object from the second range for each frame of the second image, and resets the interval between the first range and the second range when the first movement line and the second movement line cannot be combined into a single movement line.
[0079] This allows the range used to generate the target flow line to be appropriately set.
[0080] (12) The computer program of the present disclosure is a computer program executed by a setting device that is connected to multiple cameras that capture different areas of a single space and uses images obtained from the multiple cameras to set the range of images to be used to generate a target traffic line, and causes the setting device to execute the setting method described in any one of (1) to (10).
[0081] This allows the range used to generate the target flow line to be appropriately set.
[0082] The setting method, setting device, and computer program according to all claims of the present disclosure are realized by hardware resources, such as a processor, a memory, and cooperation with a computer program.
[0083] The setting method, setting device, and computer program of the present disclosure are useful for generating a movement line of an object using images captured by multiple cameras.
Claims
1. A setting method executed in a setting device connected to multiple cameras capturing different areas of a space, the setting device using images obtained from the multiple cameras to set a range of images to be used for generating a traffic line of an object, the setting device: acquiring an image including multiple frames captured by the multiple cameras; extracting a first range from each frame of a first image captured by a first camera; extracting a second range different from the first range from each frame of a second image captured by a second camera capturing an area adjacent to the first camera; detecting a first traffic line of the object from the first range for each frame of the first image; detecting a second traffic line of the object from the second range for each frame of the second image; and when the first traffic line and the second traffic line cannot be combined into a single traffic line, re-setting the distance between the first range and the second range.
2. The setting method according to claim 1, wherein the setting device sets the distance between the first range and the second range to be narrowed when the first flow line and the second flow line cannot be combined into a single flow line.
3. The setting method according to claim 1, wherein the setting device sets the distance between the first range and the second range to be wider when the first traffic line and the second traffic line overlap.
4. The setting method according to claim 1 or 2, wherein, when the first flow line and the second flow line are combined, the setting device determines the first range determined in the previous processing as a new first range, and sets the second range determined in the previous processing as a new second range.
5. The setting method according to claim 2, wherein when the setting device sets the interval between the first range and the second range to be narrowed, the size of at least one of the first range or the second range is increased.
6. The setting method according to claim 3, wherein when the setting device sets the interval between the first range and the second range to be wider, the size of at least one of the first range or the second range is reduced.
7. The setting method according to claim 2, wherein when the setting device sets the distance between the first range and the second range to be narrowed, at least one of the first range or the second range is slid closer to the other range or so as to partially overlap.
8. The setting method according to claim 3, wherein the setting device is set to slide at least one of the first range and the second range so as to approach the other range when the distance between the first range and the second range is widened.
9. The setting method according to claim 1, wherein the target is a person.
10. The setting method according to claim 1, wherein the image is an image obtained by an omnidirectional camera.
11. A setting device connected to multiple cameras capturing images of different areas of a space, the setting device using images obtained from the multiple cameras to set a range of images to be used for generating a movement line of an object, the setting device comprising an arithmetic circuit, the arithmetic circuit: acquiring images including multiple frames captured by the multiple cameras; extracting a first range from each frame of a first image captured by a first camera; extracting a second range different from the first range from each frame of a second image captured by a second camera capturing an area adjacent to the first camera; detecting a first movement line of the object from the first range for each frame of the first image; detecting a second movement line of the object from the second range for each frame of the second image; and resetting the interval between the first range and the second range when the first movement line and the second movement line cannot be combined into a single movement line.
12. A computer program executed on a setting device that is connected to multiple cameras that capture different areas of a space and uses images obtained from the multiple cameras to set the range of images to be used for generating target traffic lines, the computer program causing the setting device to execute the setting method described in claim 1.
Citation Information
Patent Citations
Monitoring device and control program of monitoring device
JP2015032069A
Display control device and method for controlling display control device
JP2015056152A
Tracking processing device
JP2016119627A
Position identification device, position identification program, route specification device, and route specification program
JP2017182761A
Tracking device, tracking method, and recording medium
WO2021245749A1