Video monitoring apparatus, video monitoring system, video monitoring method, and storage device storing video monitoring program

Abstract

An image monitoring apparatus (100) includes an acquisition unit (110) that acquires position information indicating a position of a moving body; an instruction unit (120) that instructs a photographing direction of a plurality of movable cameras (401,...) provided at predetermined positions; an evaluation unit (130) that evaluates an easily observable degree of the moving body in images captured by the plurality of movable cameras based on the position of the moving body and positions of the plurality of movable cameras; a selection unit (140) that selects a movable camera used in photographing of the moving body from the plurality of movable cameras based on the easily observable degree; and a display control unit (150) that causes a monitor screen to display an image of the moving body captured by the selected movable camera, the evaluation unit evaluating the easily observable degree based on at least one of an angular velocity of rotation of each of the plurality of movable cameras required for each of the plurality of movable cameras to continuously capture the moving body and an occlusion occurrence time in which each of the plurality of movable cameras tracks the moving body for a predetermined time.

Description

Technical Field

[0001] This invention relates to an image monitoring device, an image monitoring system, an image monitoring method, and a storage device storing an image monitoring program. Background Technology

[0002] A system has been proposed that selects a camera from multiple cameras that captures an image suitable for image processing, and performs image processing on the image captured by the selected camera. See, for example, Patent Document 1.

[0003] Existing technical documents

[0004] Patent documents

[0005] Patent Document 1: Japanese Patent Application Publication No. 2013-223104 Summary of the Invention

[0006] The problem that the invention aims to solve

[0007] However, the existing system described above cannot automatically select a PTZ (Pan-Tilt-Zoom) camera that can properly and continuously capture moving objects from among multiple movable cameras that can change shooting direction.

[0008] The purpose of this invention is to provide an image monitoring device, image monitoring system, image monitoring method, and image monitoring program that can automatically select a movable camera from a plurality of movable cameras capable of appropriately and continuously capturing moving objects.

[0009] Methods for solving problems

[0010] The image monitoring device of the present invention is characterized by comprising: an acquisition unit that acquires position information indicating the position of a moving object, i.e., a moving body; an indication unit that indicates the shooting direction of a plurality of movable cameras disposed at predetermined positions; an evaluation unit that evaluates the observability of the moving body in images captured by each of the plurality of movable cameras based on the position of the moving body and the positions of the plurality of movable cameras; a selection unit that selects a movable camera used for capturing the moving body from the plurality of movable cameras based on the observability; and a display control unit that causes a monitor screen to display the image of the moving body captured by the selected movable camera, wherein the evaluation unit evaluates the observability based on the angular velocity of rotation of each of the plurality of movable cameras required for each of the plurality of movable cameras to continuously capture the moving body.

[0011] The image monitoring method of the present invention is a computer-executed method, characterized by comprising the following steps: obtaining position information representing the position of a moving object, i.e., a moving body; indicating the shooting direction of a plurality of movable cameras set at predetermined positions; evaluating the observability of the moving body in the images captured by each of the plurality of movable cameras based on the position of the moving body and the positions of the plurality of movable cameras; selecting a movable camera from the plurality of movable cameras for use in capturing the moving body based on the observability; and displaying the image of the moving body captured by the selected movable camera on a monitor screen, wherein the observability is evaluated based on the angular velocity of rotation of each of the plurality of movable cameras required for each of the plurality of movable cameras to continuously capture the moving body.

[0012] Invention Effects

[0013] According to the present invention, it is possible to automatically select from a plurality of movable cameras that can appropriately and continuously capture moving objects. Attached Figure Description

[0014] Figure 1 This is a graph representing an example of observability.

[0015] Figure 2 This is a diagram illustrating an example of the hardware structure of the image monitoring device and image monitoring system according to Embodiment 1.

[0016] Figure 3 This diagram illustrates an example of applying the image surveillance system of Implementation 1 to the surveillance of aircraft operating at an airport.

[0017] Figure 4 This is a functional block diagram that roughly represents the structure of the image monitoring device and image monitoring system of Embodiment 1.

[0018] Figure 5 It is a graph representing the degree of following as an example of observability.

[0019] Figure 6 (A) represents an example where the rotational angular velocity is fast enough and the following accuracy is high. Figure 6 (B) represents an example of slow angular velocity and low following degree.

[0020] Figure 7 This is a flowchart illustrating the operation of the image monitoring device in Embodiment 1.

[0021] Figure 8 This diagram illustrates an example of applying the image surveillance system of Implementation 2 to the surveillance of aircraft operating at an airport.

[0022] Figure 9This is a functional block diagram that roughly represents the structure of the image monitoring device and image monitoring system of Embodiment 2.

[0023] Figure 10 This is a flowchart illustrating the operation of the image monitoring device in Embodiment 2.

[0024] Figure 11 This is a graph representing the feature degree as an example of observability.

[0025] Figure 12 This is a functional block diagram that roughly illustrates the structure of the image monitoring device and image monitoring system of Embodiment 3.

[0026] Figure 13 This is a flowchart illustrating the operation of the image monitoring device in Embodiment 3.

[0027] Figure 14 This diagram illustrates the viewpoint designation operation performed by the image monitoring system of Embodiment 3.

[0028] Figure 15 This is a diagram showing the viewpoint designation operation performed by the image monitoring system of Embodiment 3 and the images captured by the PTZ camera.

[0029] Figure 16 This diagram illustrates an example of applying the image surveillance system of Implementation 4 to the surveillance of aircraft operating at an airport.

[0030] Figure 17 This is a functional block diagram that roughly illustrates the structure of the image monitoring device and image monitoring system of Embodiment 4.

[0031] Figure 18 This is a flowchart illustrating the operation of the image monitoring device in Embodiment 4. Detailed Implementation

[0032] Hereinafter, the image monitoring device, image monitoring system, image monitoring method, and image monitoring program according to embodiments will be described with reference to the accompanying drawings. The following embodiments are merely examples, and the embodiments can be appropriately combined and modified.

[0033] The image monitoring device of the present invention can evaluate the "observability" of the moving object in the images captured by each of the multiple movable cameras based on the position of the moving object (i.e., the monitored object) and the positions of multiple movable cameras set at predetermined positions, and automatically select a movable camera from the multiple movable cameras to be used in capturing the moving object based on the observability. The movable camera is a camera capable of changing its shooting direction, such as a PTZ camera. The movable camera is also referred to as a rotating camera.

[0034] Figure 1 This is a diagram illustrating examples of observability. Observability refers to how easily a monitor (also called a "user") can observe an image of a moving object captured by a moving camera on a monitor screen. For example... Figure 1 As shown, observability can be evaluated based on one or more of the following: tracking accuracy, occlusion occurrence, detail, and eigenvalue. The higher the observability, the better the monitor can monitor moving objects in the image displayed on the monitor screen.

[0035] like Figure 1 As shown, when the angular velocity of the movable camera is fast enough (i.e., when the movable camera can continuously capture the moving object), the tracking accuracy is high; when the angular velocity of the movable camera is slow (i.e., when the movable camera cannot continuously capture the moving object, and part or all of the moving object is missing from the image), the tracking accuracy is low. High tracking accuracy results in high observability, while low tracking accuracy results in low observability. "The movable camera continuously captures the moving object" means that the orientation (i.e., the shooting direction) of the rotatable movable camera is continuously facing the moving object. "Rotation" includes horizontal rotation (translation) and vertical rotation (tilt).

[0036] like Figure 1 As shown, the occlusion occurrence rate is low when no occlusion occurs after tracking a moving object for a predetermined time (i.e., when there is no obstruction between the movable camera and the moving object). Conversely, the occlusion occurrence rate is high when occlusion occurs after tracking a moving object for a predetermined time (i.e., when there is an obstruction between the movable camera and the moving object, creating a blind spot). Furthermore, the longer the occlusion occurs, the higher the occlusion occurrence rate. Low occlusion occurrence rate results in high observability, while high occlusion occurrence rate results in low observability.

[0037] like Figure 1 As shown, when the distance from the movable camera to the moving object is short (i.e., the moving object is captured as large), the level of detail is high; when the distance from the movable camera to the moving object is long (i.e., the moving object is captured as small), the level of detail is low. High level of detail results in high observability, while low level of detail results in low observability.

[0038] like Figure 1As shown, the feature score is determined based on the direction from which the movable camera captures the moving object. High feature score occurs when the movable camera's shooting direction is one from which the characteristics of the moving object can be easily understood (e.g., shooting from a predetermined desired shooting direction for each type of moving object). Low feature score occurs when the movable camera's shooting direction is one from which the characteristics of the moving object are not easily understood (e.g., shooting from a direction different from the predetermined desired shooting direction). High feature score results in high observability, while low feature score results in low observability. Furthermore, the desired shooting direction can also be a direction specified by the user's operation.

[0039] In Implementation 1, an example of evaluating observability based on followability is described.

[0040] In Implementation 2, an example of evaluating observability based on occlusion occurrence is described.

[0041] In Implementation 3, an example of evaluating observability based on a combination of followability and featureability is described.

[0042] In Implementation 4, an example of evaluating observability based on a combination of followability and detail is described.

[0043] However, observability can also be evaluated based on other combinations of following degree, occlusion occurrence, detail, and feature. Specifically, observability can be evaluated based on any combination of following degree and occlusion occurrence, following degree and detail and feature, occlusion occurrence and detail, occlusion occurrence and feature, occlusion occurrence and detail and feature, following degree and occlusion occurrence and feature, and any combination of following degree and occlusion occurrence, detail, and feature.

[0044] Implementation Method 1

[0045] Figure 2 This diagram illustrates the hardware structure of the image monitoring device 100 and the image monitoring system 10 according to Embodiment 1. The image monitoring system 10 includes the image monitoring device 100, a monitoring panel 200, one or more fixed cameras 300 for wide-area shooting, multiple PTZ cameras 401 for narrow-area shooting, ... and a radar 500 for detecting the position of moving objects. The monitoring panel 200 includes a wide-area monitor screen 210, a narrow-area monitor screen 220, and a user operation unit 230 (described later). Figure 4 (As shown). The monitoring panel 200 may also be part of the video monitoring device 100.

[0046] The video monitoring device 100 has processing circuitry. The video monitoring device 100 is, for example, a computer. Figure 2 In the example, the video monitoring device 100 includes a CPU (Central Processing Unit) 901 as an information processing unit, a memory 902 as a program storage device, a non-volatile storage device 903 such as a hard disk drive (HDD) or solid-state drive (SSD), and an interface 904. The memory 902 is, for example, a non-volatile or volatile semiconductor memory such as RAM (Random Access Memory), ROM (Read Only Memory), flash memory, EPROM (Erasable Programmable Read Only Memory), or EEPROM (Electrically Erasable Programmable Read-Only Memory).

[0047] The memory 902 stores, for example, an image monitoring program as software. The CPU 901 can execute the image monitoring method of Embodiment 1 by executing the image monitoring program. In addition, the image monitoring device 100 may also implement part of its function through dedicated hardware and another part of its function through software or firmware.

[0048] Figure 3 This diagram illustrates an example of applying the image monitoring system 10 of Embodiment 1 to the monitoring of aircraft traveling on airport 600. The image monitoring system 10 includes an image monitoring device 100, a monitoring panel 200, fixed cameras 300, multiple PTZ cameras (#1 to #3) 401 to 403, and a radar 500 for detecting the position of aircraft M1, M2, etc., which are moving objects traveling on airport 600. The image monitoring device 100 displays images from the multiple fixed cameras 300 or composite images of the multiple fixed cameras 300 on a wide-area monitor screen 210.

[0049] Furthermore, the image monitoring device 100 obtains position information indicating the position of the aircraft M1, which is a moving monitored object, based on the detection signal from the radar 500. It then instructs the shooting directions D1, D2, and D3 of the PTZ cameras (#1 to #3) 401 to 403, which are positioned at predetermined locations. Based on the position of the aircraft M1 and the positions of the PTZ cameras (#1 to #3) 401 to 403, the image monitoring device 100 evaluates (i.e., calculates) the observability of the moving object in the images captured by each of the PTZ cameras (#1 to #3) 401 to 403. Based on the observability, the image monitoring device 100 selects the PTZ camera (#1 to #3) 401 to 403 to be used for capturing images of the aircraft M1, and displays the images of the aircraft M1 captured by the selected PTZ camera on the narrow-area monitor screen 220. The moving object, the aircraft, is, for example, an aircraft designated by the monitor through the user operation unit 230.

[0050] In Implementation 1, the ease of observation is evaluated based on the angular velocity of the rotation of each of the PTZ cameras (#1 to #3) 401 to 403, which is required for each of the PTZ cameras (#1 to #3) 401 to 403 to continuously capture the moving object.

[0051] Figure 4 This is a functional block diagram that schematically illustrates the structure of the image monitoring device 100 and the image monitoring system 10 according to Embodiment 1. The image monitoring device 100 includes: an acquisition unit 110 that acquires position information indicating the position of a moving body, i.e., an aircraft M1; an instruction unit 120 that indicates the shooting directions of a plurality of PTZ cameras (#1 to #3) 401 to 403 set at predetermined known positions; and an evaluation unit 130 that evaluates the observability of the aircraft M1 in the images captured by each of the plurality of PTZ cameras (#1 to #3) 401 to 403 based on the position of the aircraft M1 and the positions of the plurality of PTZ cameras (#1 to #3) 401 to 403. In addition, the image monitoring device 100 includes: a selection unit 140, which selects the PTZ camera used for shooting the aircraft M1 from a plurality of PTZ cameras (#1 to #3) 401 to 403 based on ease of observation; and a display control unit 150, which displays the image of the aircraft M1 captured by the selected PTZ camera on the narrow-area monitor screen 220.

[0052] The acquisition unit 110 acquires the position information of the aircraft M1 based on the detection signal of the radar 500, but it can also detect the position of moving objects based on images from fixed cameras 300 or composite images from multiple fixed cameras 300 (e.g., panoramic images or overhead images). In Embodiment 1, the evaluation unit 130 evaluates the ease of observation based on the angular velocity of the rotation of each of the multiple PTZ cameras (#1 to #3) 401 to 403, which are required to continuously capture moving objects. The evaluation unit 130 determines that the smaller the angular velocity of rotation (i.e., the higher the tracking accuracy), the higher the ease of observation (i.e., the easier it is for the observer to observe the monitored object).

[0053] Figure 5 This is a graph representing the degree of following, an example of observability. PTZ cameras (#1-#3) 401-403 rotate in correspondence with the distance (i.e., speed) traveled per unit time by the monitored moving object, aircraft M1. Figure 5 In the example, PTZ camera #1 has a small rotational angular velocity. Figure 5 In the example, PTZ camera #2 has a large rotational angular velocity. Figure 5 In the example, the rotational angular velocity of PTZ camera #3 is moderate. Therefore, the selection unit 140 sends the image of PTZ camera (#1) 401 with the smallest rotational angular velocity to the display control unit 150, and displays the image of PTZ camera (#1) 401 on the narrow-area monitor screen 220.

[0054] Figure 6 (A) represents an example where the PTZ camera (#1) 401 has a sufficiently fast rotational angular velocity and high tracking accuracy. Figure 6 (B) represents an example of a PTZ camera (#2) 402 with a slow rotational angular velocity and low tracking accuracy. For example... Figure 6 As shown in (A), in Figure 5 In the images captured by the PTZ camera (#1) 401, the PTZ camera (#1) 401 captures the entire aircraft M1 as a moving body at time t1, and continues to capture the entire aircraft M1 even after a certain time has passed to time t2. However, as Figure 6 As shown in (B), Figure 5 The PTZ camera (#2) 402 captures the entire aircraft M1 at time t1, but the tracking of the PTZ camera (#2) 402 is not timely. After a certain period of time, at time t2, it becomes unable to capture a part of the aircraft M1.

[0055] Figure 7This is a flowchart illustrating the operation of the image monitoring device in Embodiment 1. First, the acquisition unit 110 acquires location information indicating the position of the moving monitored object, i.e., the moving body (in Embodiment 1, an aircraft traveling at an airport) (step S11).

[0056] The evaluation unit 130 calculates the temporal positional relationship between the PTZ camera and the moving body (step S13), and performs a process to calculate the rotational angular velocity of the PTZ camera required for continuous capture of the moving body for all multiple PTZ cameras (step S14) (steps S12 to S15). Here, "temporal positional relationship" refers to the positional relationship between the PTZ camera and the moving body from a certain point in time to a point in time after a certain time has elapsed.

[0057] The evaluation unit 130 determines that the PTZ camera with the smaller rotational angular velocity has higher observability (step S16). The instruction unit 120 directs the PTZ camera selected based on observability to point at the moving object being monitored (step S17). At this time, the instruction unit 120 selects, for example, the PTZ camera with the highest observability.

[0058] As explained above, if the image monitoring device 100 or image monitoring method of Embodiment 1 is used, the temporal positional relationship between the moving body and the PTZ camera can be predicted. Based on this positional relationship, the image that is easiest for the monitor to observe is selected from the multiple images captured by the multiple PTZ cameras, which are multiple narrow-area monitoring cameras, and displayed on the narrow-area monitor screen 220.

[0059] Furthermore, since the PTZ camera is selected based on the temporal positional relationship between the PTZ camera and the moving object, the phenomenon of frequent image switching, i.e., zapping, displayed on the narrow-area monitor screen 220 is less likely to occur.

[0060] Implementation Method 2

[0061] Implementation 2 differs from Implementation 1 in that it uses occlusion occurrence degree as observability. Otherwise, Implementation 2 is the same as Implementation 1. Therefore, in the description of Implementation 2, reference will also be made to... Figure 1 and Figure 2 .

[0062] Figure 8 This diagram illustrates an example of applying the image surveillance system of Implementation 2 to the surveillance of aircraft M1 operating at an airport. Figure 8In the example, the PTZ camera (#1) 401 tracks and rotates to follow aircraft M1 flying at the airport. However, due to obstructions such as buildings H1, obstruction occurs at time P1, preventing it from capturing images of aircraft M1. Furthermore, due to obstructions such as buildings H2, obstruction occurs at time P2, preventing it from capturing images of aircraft M1. Additionally, in Figure 8 In the example, PTZ camera (#2) 402 tracks the aircraft M1 moving through the airport, but due to obstructions such as buildings H3, an obstruction occurs for a duration P3, preventing it from capturing images of aircraft M1. For example, when P1+P2 = 30 seconds and P3 = 15 seconds, since the PTZ camera (#2) 402, with its shorter obstruction duration, is easier to observe, the image from PTZ camera (#2) 402 is automatically selected and displayed on the narrow-area monitor screen 220.

[0063] Figure 9 This is a functional block diagram that schematically illustrates the structure of the image monitoring device 100a and the image monitoring system 10a according to Embodiment 2. Figure 9 In the middle, to and Figure 4 The constituent elements shown are the same as or correspond to the constituent element labels. Figure 4 The labels shown are the same.

[0064] The video surveillance system 10a includes a video surveillance device 100a, a monitoring panel 200, one or more fixed cameras 300, multiple PTZ cameras (#1 to #3) 401 to 403, and a radar 500 for detecting the position of an aircraft M1 moving at the airport. The video surveillance device 100a displays the images from the fixed cameras 300 or a composite image of the images from multiple fixed cameras 300 on a wide-area monitor screen 210.

[0065] Furthermore, the image monitoring device 100a acquires position information indicating the position of the aircraft M1 moving on the taxiway, runway, etc. at the airport via radar 500, instructs the shooting direction of PTZ cameras (#1 to #3) 401 to 403 set at predetermined positions, evaluates the observability of the aircraft M1 in the images captured by each of the PTZ cameras (#1 to #3) 401 to 403 based on the observability, selects the PTZ camera to be used for capturing the aircraft M1 from the PTZ cameras (#1 to #3) 401 to 403, and displays the image of the aircraft M1 captured by the selected PTZ camera on the narrow-area monitor screen 220. Additionally, the moving object, the aircraft M1, is specified by the monitor, for example, through the user operation unit 230.

[0066] In embodiment 2, the ease of observation is evaluated based on the occlusion occurrence degree of each of the PTZ cameras (#1 to #3) 401 to 403. The evaluation unit 130a of the image monitoring device 100a calculates the occlusion occurrence time based on the position information of the PTZ cameras and the occlusion information, and selects the image to be displayed on the narrow-area monitor screen 220 based on the occlusion occurrence degree corresponding to the occlusion occurrence time.

[0067] Figure 10 This is a flowchart illustrating the operation of the image monitoring device 100a in Embodiment 2. First, the acquisition unit 110 acquires location information indicating the position of the moving monitored object, i.e., the moving body (in Embodiment 2, an aircraft traveling at an airport) (step S21).

[0068] Evaluation unit 130a calculates the temporal positional relationship between the PTZ camera and the moving body (step S23), and performs a process to calculate the occlusion occurrence time within a predetermined time for all multiple PTZ cameras (step S24) (steps S22 to S25).

[0069] Evaluation unit 130a determines that the PTZ camera with a lower occlusion occurrence degree corresponding to the occlusion occurrence time has higher observationability (step S26). Instruction unit 120 directs the PTZ camera selected based on observationability to point at the moving object being monitored (step S27). At this time, instruction unit 120 selects, for example, the PTZ camera with the highest observationability.

[0070] As explained above, if the image monitoring device 100a or image monitoring method of Embodiment 2 is used, the image that is easiest for the monitor to observe can be displayed on the narrow-area monitor screen 220 from the multiple images captured by the multiple PTZ cameras that are multiple narrow-area monitoring cameras.

[0071] Furthermore, since the PTZ camera is selected based on the temporal positional relationship between the PTZ camera and the moving object, the phenomenon of frequent image switching, i.e., zapping, displayed on the narrow-area monitor screen 220 is less likely to occur.

[0072] Implementation Method 3

[0073] The difference between Embodiment 3 and Embodiment 1 is that Followability and Feature Degree are used as observability. Otherwise, Embodiment 3 is the same as Embodiment 1. Therefore, in the description of Embodiment 3, reference is also made to... Figure 1 and Figure 2 .

[0074] Figure 11 This is a graph representing an example of characteristics as observability. In Figure 11In the example, when a vector V with a desired shooting direction is specified (e.g., when vector V is specified by user operation or when vector V is predetermined), the evaluation unit automatically selects a PTZ camera with a shooting direction similar to that vector V, and displays the image of the automatically selected PTZ camera on the narrow-field monitor screen 220. Figure 11 In the example, the feature degree is calculated as the similarity between the vector V of the desired shooting direction and the vectors V1 to V4 of the shooting directions of each PTZ camera. Figure 11 In the example, since vector V3 among the vectors V1 to V4 of the shooting direction is most similar to the vector V of the desired shooting direction, PTZ camera #3 403 is automatically selected. Vector V can also be determined in advance based on the type of moving object. Alternatively, vector V can be determined based on instructions input by the monitor.

[0075] Figure 12 This is a functional block diagram that schematically illustrates the structure of the image monitoring device 100b and the image monitoring system 10b in Embodiment 3. Figure 12 In the middle, to and Figure 4 The constituent elements shown are the same as or correspond to the constituent element labels. Figure 4 The labels shown are the same. Figure 12 In the image monitoring device 100b shown, the evaluation unit 130b calculates the tracking degree and feature degree based on the position information and viewpoint designation information of the PTZ camera, and selects the image to be displayed on the narrow-area monitor screen 220 based on the comprehensive evaluation value of the combination of tracking degree and feature degree. The comprehensive evaluation value is calculated, for example, by weighted summation of tracking degree and feature degree. When the weight coefficient of tracking degree is set to w1 and the weight coefficient of feature degree is set to w2, the comprehensive evaluation value E is calculated by the following formula.

[0076] E = (Following degree) × (w1) + (Feature degree) × (w2)

[0077] Figure 13 This is a flowchart illustrating the operation of the image monitoring device 100b in Embodiment 3. First, the acquisition unit 110 acquires position information indicating the position of the moving object, i.e., the moving body (in Embodiment 3, an aircraft traveling at an airport) (step S31).

[0078] Evaluation unit 130b calculates the temporal positional relationship between the PTZ camera and the moving object (step S33), calculates the rotational angular velocity of the PTZ camera required to continuously capture the moving object (step S34), calculates the feature degree of the moving object (step S35), and performs processing to calculate the comprehensive evaluation value for all multiple PTZ cameras (step S36) (steps S32 to S37).

[0079] The evaluation unit 130b determines that the higher the overall evaluation value of the PTZ camera, the higher its ease of observation (step S38). The instruction unit 120 directs the PTZ camera selected based on its ease of observation to point at the moving object being monitored (step S39). At this time, the instruction unit 120 selects, for example, the PTZ camera with the highest ease of observation.

[0080] Figure 14 This diagram illustrates the operation of specifying the viewpoint in the image monitoring system 10b according to Embodiment 3. By dragging the narrow area of ​​the monitor screen 220 displayed on the monitoring panel 200b, the three-dimensional model (i.e., 3D model) of the aircraft, viewed from a diagonal front on the image, is rotated, and the rear of the 3D model of the aircraft is displayed on the image, thereby inputting the vector V of the shooting direction. This operation can also be performed via a user operation unit 230b such as a mouse, switch, or touch panel.

[0081] Figure 15 This is a diagram showing the viewpoint designation operation performed by the image monitoring system 10b of Embodiment 3 and the images captured by the PTZ camera. Figure 15 The following scenario is illustrated: Dragging the display of the narrow-field monitor screen 220 rotates the 3D model of the aircraft on the image, displaying the rear of the 3D model of the aircraft as an image, thereby inputting the shooting direction vector V. Before performing a viewpoint change operation from the viewpoint change interface, the image of PTZ camera #3 is displayed on the narrow-field monitor screen 220; however, after performing a viewpoint change operation from the viewpoint change interface, the image of PTZ camera #1 is displayed on the narrow-field monitor screen 220.

[0082] As explained above, if the image monitoring device 100b or image monitoring method of Embodiment 3 is used, the image that is easiest for the monitor to observe can be automatically displayed on the narrow-area monitor screen 220 from the multiple images captured by the multiple PTZ cameras that are multiple narrow-area monitoring cameras.

[0083] In addition, the viewpoint change interface allows for intuitive and easy-to-understand operations such as rotating the 3D model of the monitored object, i.e., the aircraft, to display a specified image on the narrow-area monitor screen 220.

[0084] Implementation Method 4

[0085] The difference between Embodiment 4 and Embodiment 1 is that followability and detail are used as observability. Otherwise, Embodiment 4 is the same as Embodiment 1. Therefore, in the description of Embodiment 4, reference is also made to... Figure 1 and Figure 2 .

[0086] Figure 16This diagram illustrates an example of applying the video surveillance system of Embodiment 4 to the surveillance of an aircraft M1 traveling at an airport. PTZ cameras (#1 to #3) 401 to 403 rotate in accordance with the distance (i.e., speed) traveled per unit time by the monitored moving object, namely the aircraft M1. Figure 16 In the example, the PTZ camera (#1) 401 has a small rotational angular velocity. Additionally, in Figure 16 In the example, the PTZ camera (#2) 402 has a large rotational angular velocity. Additionally, in... Figure 16 In the example, the PTZ camera (#3) 403 has a moderate rotational angular velocity. Additionally, in Figure 16 In the example, the distance from PTZ camera (#1) 401 to aircraft M1 is 300m, the distance from PTZ camera (#2) 402 to aircraft is 300m, and the distance from PTZ camera (#3) 403 to aircraft M1 is 600m. As a comprehensive evaluation value obtained by weighted summation of these values, PTZ camera (#1) 401 has the highest value. Therefore, the selection unit 140 sends the image of PTZ camera (#1) 401 with the smallest rotational angular velocity to the display control unit 150, and displays the image of PTZ camera (#1) 401 on the narrow-area monitor screen 220.

[0087] Figure 17 This is a functional block diagram that schematically illustrates the structure of the image monitoring device 100c and the image monitoring system 10c in Embodiment 4. Figure 17 In the middle, to and Figure 4 The constituent elements shown are the same as or correspond to the constituent element labels. Figure 4 The labels shown are the same. Figure 17 In the image monitoring device 100c shown, the evaluation unit 130c calculates the tracking degree and detail based on the position information of the PTZ camera, and selects the image to be displayed on the narrow-area monitor screen 220 based on the comprehensive evaluation value of the combination of tracking degree and detail.

[0088] Figure 18 This is a flowchart illustrating the operation of the image monitoring device in Embodiment 4. First, the acquisition unit 110 acquires position information indicating the position of the moving object, i.e., the moving body (in Embodiment 4, an aircraft traveling at an airport) (step S41).

[0089] Evaluation unit 130c calculates the temporal positional relationship between the PTZ camera and the moving object (step S43), calculates the rotational angular velocity of the PTZ camera required to continuously capture the moving object (step S44), calculates the detail of the moving object (step S45), and performs processing to calculate the comprehensive evaluation value for all multiple PTZ cameras (step S46) (steps S42 to S47).

[0090] The evaluation unit 130c determines that the higher the overall evaluation value of the PTZ camera, the higher its ease of observation (step S48). The instruction unit 120 directs the PTZ camera selected based on its ease of observation to point at the moving object being monitored (step S49). At this time, the instruction unit 120 selects, for example, the PTZ camera with the highest ease of observation.

[0091] As explained above, if the image monitoring device 100c or image monitoring method of Embodiment 4 is used, the image that is easiest for the monitor to observe can be automatically displayed on the narrow-area monitor screen 220 from the multiple images captured by the multiple PTZ cameras that are multiple narrow-area monitoring cameras.

[0092] Variations

[0093] The above description refers to an image surveillance system that monitors aircraft as the moving object and monitors aircraft flying at an airport. However, the moving object is not limited to aircraft. The moving object can also be a car traveling on a road, a ship moving on water, an aircraft flying in the air, a person walking on a road, etc.

[0094] In the above description, the ease of observation is determined based on a weighted sum of one or more values ​​of following degree, occlusion occurrence degree, detail degree, and feature degree. The PTZ camera used for tracking moving objects is then selected based on the determined ease of observation. However, the video surveillance device may also be configured to change which of the following—following degree, occlusion occurrence degree, detail degree, feature degree, or a combination thereof—is used to determine ease of observation based on user operation or the type of moving object. For example, the video surveillance device may evaluate ease of observation based on occlusion occurrence degree when the moving object being monitored is an aircraft operating at an airport, and based on following degree when the moving object being monitored is an aircraft flying in the air.

[0095] The above description illustrates an example of a video surveillance system with one narrow-area monitor screen 220, but it can also have multiple narrow-area monitor screens 220. In this case, the images can be displayed sequentially, starting with those that are easiest to observe.

[0096] Label Explanation

[0097] 10, 10a-10c: Video surveillance system; 100, 100a-100c: Video surveillance device; 110: Acquisition unit; 120: Indication unit; 130, 130a-130c: Evaluation unit; 140: Selection unit; 150: Display control unit; 160: Viewpoint designation unit; 200, 200b: Monitoring panel; 210: Wide-area monitor screen; 220: Narrow-area monitor screen (monitor screen); 230, 230b: User operation unit; 300: Fixed camera; 401-403: PTZ camera (movable camera); 500: Radar; 600: Airport; D1-D3: Shooting direction; M1, M2: Aircraft (moving body); H1-H3: Obstructions.

Claims

1. A video monitoring device, characterized in that, The video monitoring device has the following features: The acquisition unit acquires position information indicating the position of the moving object, i.e., the moving body. An indicator unit that indicates the shooting direction of multiple movable cameras set in a predetermined position; The evaluation unit evaluates the observability of the moving object in the images captured by each of the plurality of movable cameras, based on the position of the moving object and the position of the plurality of movable cameras. The selection unit selects, based on the ease of observation, a movable camera used for filming the moving object from among the plurality of movable cameras; as well as The display control unit causes the monitor screen to display the image of the moving object captured by the selected movable camera. The evaluation unit evaluates the ease of observation based on the angular velocity of rotation of each of the plurality of movable cameras required for each of the plurality of movable cameras to continuously capture the moving object. The smaller the angular velocity, the higher the observability. The selection unit selects a movable camera with high visibility.

2. The image monitoring device according to claim 1, characterized in that, The evaluation unit evaluates the observability of the moving object in the images captured by each of the plurality of movable cameras based on the angular velocity and the distance between the position of the moving object and the positions of the plurality of movable cameras.

3. The image monitoring device according to claim 1, characterized in that, The evaluation unit evaluates the observability of the moving object in the images captured by the plurality of movable cameras based on the angular velocity and the feature density of the moving object contained in the images captured by the plurality of movable cameras. The feature degree is calculated as the similarity between the vector of the desired shooting direction and the vector of the shooting direction of each of the plurality of movable cameras.

4. The image monitoring device according to claim 1, characterized in that, The evaluation unit evaluates the observability of the moving object in the images captured by each of the multiple movable cameras based on the angular velocity, the distance between the position of the moving object and the positions of the multiple movable cameras, and the feature density of the moving object contained in the images captured by each of the multiple movable cameras. The characteristic is determined based on a predetermined portion of the moving body or a portion specified by the user.

5. The image monitoring device according to claim 1, characterized in that, The evaluation unit evaluates the observability based on the angular velocity and the occlusion occurrence degree corresponding to the occlusion occurrence time that occurs after the plurality of movable cameras have each tracked the moving body for a predetermined time. The smaller the angular velocity and the smaller the degree of occlusion, the higher the observability.

6. The image monitoring device according to claim 5, characterized in that, The evaluation unit evaluates the observability of the moving object in the images captured by each of the plurality of movable cameras based on the angular velocity, the degree of occlusion, and the distance between the position of the moving object and the positions of the plurality of movable cameras.

7. The image monitoring device according to claim 5, characterized in that, The evaluation unit evaluates the observability of the moving object in the images captured by the plurality of movable cameras based on the angular velocity, the occlusion occurrence degree, and the feature degree of the moving object contained in the images captured by the plurality of movable cameras. The feature degree is calculated as the similarity between the vector of the desired shooting direction and the vector of the shooting direction of each of the plurality of movable cameras.

8. The image monitoring device according to claim 5, characterized in that, The evaluation unit evaluates the observability of the moving object in the images captured by each of the multiple movable cameras based on the angular velocity, the degree of occlusion, the distance between the position of the moving object and the positions of the multiple movable cameras, and the feature density of the moving object contained in the images captured by each of the multiple movable cameras. The feature degree is calculated as the similarity between the vector of the desired shooting direction and the vector of the shooting direction of each of the plurality of movable cameras.

9. The image monitoring device according to any one of claims 3, 7 or 8, characterized in that, The evaluation unit determines the desired shooting direction based on the operation of the user operation unit, which has a display function that displays a model simulating the moving object and an operation function that rotates the model through user operation, and evaluates the observability of the moving object in the images captured by each of the plurality of movable cameras.

10. A video surveillance system, characterized in that, This video surveillance system has the following features: The image monitoring device according to any one of claims 1 to 9; and The plurality of movable cameras.

11. A video surveillance method, executed by a computer, characterized in that, The image monitoring method comprises the following steps: Obtain the position information of the moving object, i.e., the position of the moving body; Indicates the shooting direction of multiple movable cameras set in predetermined locations; Based on the position of the moving object and the positions of the plurality of movable cameras, evaluate the observability of the moving object in the images captured by each of the plurality of movable cameras; Based on the ease of observation, a movable camera for use in filming the moving object is selected from the plurality of movable cameras; as well as The monitor screen displays an image of the moving object captured by the selected movable camera. The observability is evaluated based on the angular velocity of rotation of each of the plurality of movable cameras required for each of the plurality of movable cameras to continuously capture the moving object. The smaller the angular velocity, the higher the observability. Choose a movable camera that is easy to observe.

12. A storage device storing an image monitoring program, characterized in that, The image monitoring program causes the computer to execute the image monitoring method as described in claim 11.

Images