JPH10174090A - Monitor supporting device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To efficiently monitor an object which is desired to be marked by a monitoring person by providing a man-machine interface means for designating a marking area in a picture, and a frame control means for controlling a frame based on positional information in the picture of the making area specified by this means. SOLUTION: A picture picked up by an image input means 101 and graphic information generated by that the monitoring person operates the mouse of a personal computer or the like, are synthesized by a synthesizing means 103 inside the personal computer. This is displayed in a picture display means 104. The monitor moves a mouse cursor to an area which he/her wanting to mark and designates as the marking area. The positional information of the designated marking area is transmitted to the frame control means 107 via a serial interface mounted in the personal computer. The frame control means 107 drives and controls the viewing direction of the picture input means 101 so that the designated marking area is imaged at the center of the screen for example, using the positional information of the marking area.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a general monitoring operation in which an image input device such as a television camera is installed in a monitored area and a monitoring person constantly monitors a monitor in another room. The present invention relates to a monitoring support device that reduces the work load of a monitoring person and assists a monitoring person in performing monitoring work efficiently.

[0002]

2. Description of the Related Art As a conventional monitoring support apparatus, there is known a moving object tracking apparatus described in Japanese Patent Application Laid-Open No. 5-346958. This is to reduce the work load on the observer by detecting a moving object by image processing and automatically tracking the moving object.

FIG. 7 shows a configuration of a moving object tracking device described in Japanese Patent Application Laid-Open No. 5-346958. In FIG. 7, 7
01 is an image input means. A first image memory 702 stores the n-th image data output from the image input unit 701. A second image memory 703 can store the (n + 1) th image data output from the image input unit 701 and change the output start timing.
Reference numeral 704 denotes a positioning unit which detects a displacement vector of the image position between the first image memory 702 and the second image memory 703. Reference numeral 705 denotes a moving object detecting means,
The moving object is detected based on the image difference between the first image memory 702 and the second image memory 703 aligned by the step 04. Reference numeral 706 denotes a position calculating unit that obtains the center of gravity of the image of the moving object obtained by the moving object detecting unit 705, and calculates the position of the moving object with the center of the screen as the origin. Reference numeral 707 denotes a gantry, which freely drives the viewing direction of the image input unit 701. Reference numeral 708 denotes a gantry control unit that controls the gantry 707. A register 709 stores the output of the position calculation unit 706. Reference numeral 710 denotes an adding means, which outputs the output of the positioning means 704 and the register 70.
9 to control the output start timing of the second image memory 703. Reference numeral 711 denotes a control unit, which includes a first image memory 702, a second image memory 703, a positioning unit 704, a moving object detection unit 705, a position calculation unit 706,
Each processing procedure of the register 709 and the adding means 710 is controlled.

[0004] The operation of the mobile object tracking device configured as described above will be described below. FIG. 8 shows the processing procedure. First, it is assumed that the processing of the n-th frame has already been completed and the image data of the n-th frame (for example, the image of FIG. 9A) input by the image input unit 701 is stored in the first image memory 702. I do. Therefore, the (n + 1) th process is started. Now, the image data in the (n + 1) th frame input by the image input means 701 according to the instruction of the control means 711 is stored in the second image memory 703. At this time, since the field-of-view image has been panned (tilted) by the n-th process, the image is as shown in FIG. 9B (ST81).

[0005] Next, the control means 711
By clearing the output of the register 04 (shift amount x, y = 0, 0) (ST82), the output of the adding means 710 is output as it is from the register 709. Here, the register 709 stores the contents of the n-th position calculating unit 706,
That is, since the detected positions x (n) and y (n) of the moving object in the coordinate system having the origin at the center of the screen at the n-th are stored, the adding means 710 sets the second image memory 703
Output start timing of x (n) in the x direction and y in the y direction
By delaying (or moving forward) by (n), the second image memory 703 outputs an image whose position has been corrected by x (n) and y (n) (the image in FIG. 9C). At this time, the control means 711 sends the output of the first image memory 702 and the output of the second image memory 703 to the positioning means 704 (ST83). Then, the positioning means 704 detects the displacement vector of the images of FIGS. 9A and 9C (ST84).

As a method of calculating the displacement vector, for example, when the image at the n-th time is gn (x, y) and the image at the n + 1-th time is gn + 1 (x, y), the image g
n (x, y) is shifted by ξ in the x direction and η in the y direction,
Ξ and η that minimize. ΣΣ | gn (x−ξ, y−η) −gn + 1 (x, y) | (1) As a result, the displacement vectors Δx (n + 1) and Δy between the images of FIGS. 9A and 9C are obtained. (N + 1) can be detected. Therefore, the control unit 711 sets the positioning unit 70
4, ie, the displacement vector Δx (n +
1), by outputting Δy (n + 1) to the adding means 710, the adding means 710 makes the content of the register 709 = x
(N), y (n) and displacement vector Δx (n + 1),
Δy (n + 1) is added (ST85), and the output start timing of the second image memory 703 is set to x (n) + Δx in the x direction.
By delaying (or moving forward) by (n + 1) and y (n) + Δy (n + 1) in the y direction, the second image memory 703 stores x (n) + Δx (n + 1) and y (n) + Δy
An image whose position is corrected by (n + 1) is output. At this time, the control means 711 sends the output of the first image memory 702 and the output of the second image memory 703 to the moving object detection means 705 (ST86).

In this way, the moving object detecting means 705 compares the n-th image with the (n + 1) -th image whose position has been corrected by the positioning means 704. If the difference between the two images is obtained, the moving object is detected from within the image. Only can be detected (ST87).

The position calculating means 706 calculates the center of gravity of the detected moving body, calculates the positions x (n + 1) and y (n + 1) of the moving body in a coordinate system having the origin at the center position of the image. By updating the content (ST88) and outputting it to the gantry control means 708, the gantry 707 is controlled and the moving object can be captured at the center of the visual field image of the image input means 701 (ST89). The moving object can be tracked by repeating the above operation.

[0009]

However, in the above-mentioned conventional moving object tracking apparatus, only the moving object detected by the detection of the moving object by the image processing is considered as the object to be tracked to the last. This is effective when a suspicious person or the like occasionally enters the non-existent monitored area.However, in the monitored area where there are multiple moving objects, the monitor should pay special attention based on the judgment of the monitor. It is not suitable when you want to track your body. This is because the object (region) that the observer wants to pay attention to does not necessarily match the object detected by the moving object detection. As an example, let's assume that an observer finds and monitors a suspicious individual while watching the video of a surveillance camera installed at a department store's sales floor in a monitor in another room. In such a case, there are many persons (customers, etc.) in the monitored area, that is, the sales floor, and it is difficult to determine which object should be tracked in the related art. For example, it is possible to keep track of the first detected object or to preferentially track a newly detected object. However, it will be apparent at a glance that the moving object (tracking target) determined in this way does not always match the object that the observer wants to pay attention to. A skilled observer picks up a suspicious person from the individual's detailed behavior and monitors the individual's attention. At present, such surveillance work is performed by a monitor using a joystick or other remote control device in a monitor room to control the direction of view of the surveillance camera (control the gantry) or control the angle of view (zoom control). The operation requires considerable skill.

[0010] The present invention solves the above-mentioned conventional problems. Even if a plurality of moving or stationary objects exist in an image,
It is an object of the present invention to provide an excellent surveillance support device that enables a monitor to efficiently monitor a target object that the watcher wants to pay attention to and reduce the work load on the monitor.

[0011]

In order to achieve the above object, in a monitoring support apparatus according to the present invention, an observer himself / herself looks at a target area where an object of interest such as a suspicious person is present while looking at the monitor. Man-machine interface means that can be specified in the image, and gantry control means that automatically controls the direction of the field of view of the image input means in the direction of the specified object of interest (area of interest) and captures the area of interest at a specific position in the image Thus, even if a plurality of moving or stationary objects are present in the image, the monitor can efficiently monitor the target object that the user wants to pay attention to, thereby reducing the work load on the monitor.

[0012]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS A monitoring support apparatus according to a first aspect of the present invention monitors image input means, a gantry for controlling the direction of the field of view of the image input means, and an image input by the image input means. Man-machine / specify the area of interest in the image
Interface means; and gantry control means for controlling the gantry based on positional information in the image of the region of interest specified by the man-machine interface means.

According to a second aspect of the present invention, in the monitoring support apparatus according to the first aspect, the gantry control means obtains the area of interest designated by the man-machine interface means by the image input means after the gantry control. The gantry is controlled so that an image to be taken is captured at a specific position set in advance.

According to the above-described configurations, instead of tracking a moving body (and the number of moving bodies is limited to one in most cases) in which the judgment of the observer is not taken into account as in the related art, the target of interest is monitored only. By specifying in the image by the worker, only the object that the watchman really wants to pay attention to can be efficiently monitored. Therefore, since there is no need for skill like a manual operation of a stand or a zoom using a joystick or the like which has been conventionally used, there is an effect that the work load on the monitor can be greatly reduced.

According to a third aspect of the present invention, there is provided a monitoring support apparatus for monitoring an image input by an image input means, a gantry for controlling a view direction of the image input means, and monitoring an image input by the image input means. Man-machine interface means for designating an area; movement direction detection means for detecting a movement direction of an object at a specific position of a preset image by using a plurality of images inputted by the image input means; And / or gantry control means for controlling the gantry based on the moving direction.

According to a fourth aspect of the present invention, in the monitoring support apparatus according to the third aspect, the gantry control means obtains the area of interest designated by the man-machine interface means by the image input means after the gantry control. The gantry is controlled so that an image is taken at a preset specific position of the image to be taken, and a region of interest is always taken at a preset specific position of the image based on the moving direction detected by the moving direction detecting means. The gantry is controlled as described above.

According to each of the above-described configurations, instead of following a moving object in which the judgment of the observer is not taken into account as in the related art, the observer simply wants to pay attention to the target by designating the target of interest on the screen. The object can be specified, and once specified, even if the object moves, the object is automatically tracked, so that efficient monitoring can be performed. Therefore, since there is no need for skill like manual operation of a stand or an angle of view using a joystick or the like which has been conventionally used, there is an effect that the work load on the observer can be greatly reduced.

According to a fifth aspect of the present invention, there is provided a monitoring support apparatus comprising: an image input unit having a zoom function; a gantry for controlling a direction of a visual field of the image input unit; and an image input by the image input unit. Moving direction detection for detecting a moving direction of an object at a specific position of a preset image using man-machine interface means for designating a region of interest in the image and plural images inputted by the image input means; Means, gantry control means for controlling the gantry based on the region of interest and / or moving direction, and view angle control means for performing zoom control of the image input means depending on the size and shape of the region of interest. .

According to the above configuration, instead of following a moving object in which the judgment of the observer is not taken into account as in the related art, the observer specifies the target of interest in the image to the last, so that the object that the observer really wants to pay attention to can be obtained. In addition to automatically controlling the gantry so that the area of interest can be imaged at a specific position in an image in which the area of interest has been set in advance, it also automatically zooms in on the area of interest, making conventional monitoring work easier. This has the effect that close monitoring can be performed with a simpler operation.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. (Embodiment 1) FIG. 1 shows the configuration of a monitoring support apparatus according to Embodiment 1 of the present invention. In FIG.
An image input unit 101 captures an image of the monitored area and outputs the image to the outside. Reference numeral 102 denotes a man-machine interface unit for monitoring an image input by the image input unit 101 by a monitor or the like and specifying a region of interest in the image. Combining means 103 for combining the graphic information from the attention area designating means 105, and an image display means 104 for displaying the image combined by the combining means 103.
And a region-of-interest designating unit 105 that designates a region of interest while monitoring the image displayed on the image display unit 104 and outputs position information and the like of the designated region of interest to the gantry control unit 107. Reference numeral 107 denotes a gantry control unit, which controls the view direction of the image input unit 101 by driving the gantry 106 using position information and the like from the attention area specifying unit 105.

Next, the operation of the monitoring support apparatus according to the first embodiment will be described with reference to FIGS. Here, the man-machine interface means 10
As a specific example of implementation 2, a case where a personal computer is used as shown in FIG. 2 will be described as an example. Image input means 1
01 and graphic information such as a mouse cursor generated by a monitor operating a personal computer mouse or the like in a personal computer, are synthesized by a synthesizing means 103 in the personal computer, and are displayed on an image display means 104 (a personal computer monitor). Is displayed. As shown in FIG. 2 (A-1) or (A-2), the observer moves the mouse cursor to an area to which he / she wants to pay attention, and designates the area of interest. The position information of the designated region of interest is output to the gantry control unit 107 through, for example, a serial interface such as RS-232C provided in the personal computer. Mount control means 1
07, the view direction of the image input means 101 is set to the gantry 1 so that the designated region of interest is imaged, for example, at the center of the screen as shown in FIG.
06 is driven.

As described above, according to the first embodiment, even when a plurality of moving objects are present in the monitored area, not only can the monitoring person efficiently monitor the object that the user really wants to pay attention to, but also can be used conventionally. Unlike the manual operation of the stand and angle of view using a joystick or the like, skill is not required, so that the work load on the observer can be greatly reduced.

(Embodiment 2) FIG. 3 shows a configuration of a monitoring support apparatus according to Embodiment 2 of the present invention. In FIG. 3, reference numeral 301 denotes an image input unit which captures an image of a monitored area and outputs the image to the outside. 302 is a man-machine
The interface means monitors the image input by the image input means 301 by the observer and designates a region of interest within the image. The image signal from the image input means 301 and the region of interest designation means 305 described later Combining means 303 for combining graphic information from
3. An image display unit 304 for displaying the image synthesized by the third unit 3 and a switching unit 311 for specifying a region of interest while monitoring the image displayed on the image display unit 304 and switching position information and the like of the specified region of interest. And a target area designating means 305 to be output. Reference numeral 306 denotes a moving direction detecting means for detecting the moving direction of the object in the region of interest designated by the man-machine interface means 302 and automatically tracking the object. Image memory for storing each image, here, a first image memory 307 and a second image memory 308, and a movement vector detecting means 309 for detecting a movement vector of a predetermined position in the image from a plurality of images. And the movement vector detecting means 309
And a position calculating means 310 for calculating the position of the object after the movement from the movement vector detected by the method. 3
Reference numeral 11 denotes switching means, which is a man-machine interface means 30 when a region of interest is designated by a monitor.
Then, the signal is switched so that the position information from the moving direction detecting means 306 is output. Reference numeral 313 denotes a gantry control unit that controls the view direction of the image input unit 301 by driving the gantry 312 using position information and the like from the switching unit 311.

Next, the operation of the monitoring support apparatus according to the second embodiment will be described with reference to FIGS. Here, as in the first embodiment, as a specific example of the realization of the man-machine interface means 302, a case where a personal computer is used as shown in FIG. 2 or FIG. 4 will be described as an example. The image captured by the image input means 301 and graphic information such as a mouse cursor generated by a monitor operating a personal computer mouse or the like are synthesized by the synthesizing means 303 inside the personal computer,
The image is displayed on the image display means 304 (personal computer monitor).
The observer moves the mouse cursor to an area he or she wants to pay attention to, as shown in FIG. 2 (A-1) or (A-2),
Specify the region of interest. The position information of the designated area of interest is switched by the switching means 3 through, for example, a serial interface such as RS-232C provided in the personal computer.
11 is output. When the region of interest is designated, the position information is sent from the switching unit 311 to the gantry control unit 313, and the gantry control unit 313 uses the position information of the region of interest to display the designated region of interest, for example, as shown in FIG. like,
The direction of the field of view of the image input device 301 is controlled by driving the gantry 312 so that the image is captured at the center of the screen.

Thereafter, the moving direction of the region indicated by the rectangle at the center of the screen in FIG. 4A (which is set in the image in advance) is continuously detected, and the position information after the movement is used to detect the moving direction in FIG.
As shown in (B) and (C), the gantry 312 is controlled such that the object of interest is always imaged at the center of the screen. The moving direction detecting means 306 detects this moving direction. The plurality of images captured at different times by the image input unit 301 are temporarily stored in an image memory, here, the first image memory 30.
7 and the second image memory 308. Next, the movement vector detection means 309 determines whether each of the image memories 307, 3
08, a predetermined specific position in the image, for example, a movement vector at the center of the screen (the center rectangular area in FIG. 4A) is detected from the plurality of images stored in the image data 08. The position information of the target object after the movement is calculated. The position information of the object of interest after the movement calculated by the position calculating unit 310 is sent to the switching unit 311. After the attention area is specified, the switching unit 311 switches the internal circuit so that the position information output from the moving direction detection unit 306 is output to the gantry control unit 313. Therefore, since the gantry control unit 313 controls the gantry 312 based on a specific position in the image, for example, the position information after the movement of the center of the screen after the attention area is specified, the object of interest is always imaged at the center of the screen. become. In addition, as a method of detecting the movement vector, for example, the movement vector can be obtained by calculating ξ and η that minimize Expression 1.

As described above, according to the second embodiment, even when there are a plurality of moving objects in the monitored area, the observer can accurately specify the object that he / she really wants to pay attention to, and after specifying once, Even if the object moves, the object is automatically tracked, so that only the target object can be efficiently monitored. In addition, since it does not require skill like manual operation of a stand or an angle of view using a joystick or the like which has been conventionally used, the work load on a supervisor can be greatly reduced.

(Embodiment 3) FIG. 5 shows the configuration of a monitoring support apparatus according to Embodiment 3 of the present invention. The third embodiment has almost the same configuration as the second embodiment, and the operation is almost the same. The different points are an image input unit 501 and an angle of view control unit 514.
Here, these configurations and their operations will be described with reference to FIGS. In the third embodiment, the image input means 501 has a zoom function. For example, as shown in FIG. 6A, a region of interest is designated by the man-machine interface unit 502, and its position information (the coordinates of the end point of the rectangle, the position of the center of gravity, etc.) is specified.
Is sent to the angle-of-view control means 514. Angle-of-view control means 514
Then, the optimum zoom magnification is calculated from the position information, and the angle of view of the image input unit 501 is controlled. The operation described in the second embodiment (the image obtained by this operation is shown in FIG.
Shown in ) And the operation of the angle-of-view control, the monitor image in the third embodiment is as shown in FIG.

As described above, according to the third embodiment, even when there are a plurality of moving objects in the monitored area, the observer can accurately specify the object that the user really wants to pay attention to, and after specifying once, Even if the object moves, the object is automatically tracked, so that only the target object can be efficiently monitored. In addition, since the optimum angle of view control is performed based on the position information (for example, the end point of the rectangle or the position of the center of gravity) of the designated region of interest, zoom monitoring of the region of interest becomes possible, and precise monitoring can be performed with a simpler operation than before. It can be carried out. In addition, since it does not require skill like manual operation of a stand or an angle of view using a joystick or the like which has been conventionally used, the work load on a supervisor can be greatly reduced.

[0029]

As described above, claim 1 or claim 2
According to the invention described in (1), it is possible for the observer to specify the target area where the target object such as a suspicious person exists in the image while watching the monitor, and the designated target object (target area) The direction of the field of view of the image input means is automatically controlled in the direction of, and the region of interest is imaged at a specific position of the image. In most cases, the number of moving objects is limited to one. Instead of tracking the object, the observer specifies the target of interest in the image to the last, thereby efficiently monitoring only the object that the observer really wants to pay attention to. it can. Further, since skill is not required unlike a conventional operation of a gantry or the like using a joystick or the like, manual operation of a monitor can be greatly reduced.

According to the third or fourth aspect of the present invention, first, the field of view is changed so that the object of interest set by the man-machine interface means is picked up at a predetermined image specific position. After that, if the object of interest moves, the direction of movement is detected, and based on the direction of movement, the field of view is changed so that the object of interest is always captured at a preset image specific position. Since the control is continued, instead of following the moving object where the judgment of the observer is not taken into account as in the past, the observer designates the target of interest in the screen, so that the observer truly focuses on The object to be specified can be specified, and once specified, even if the object moves, the object is automatically tracked so that efficient monitoring can be performed. Similarly, since skill is not required unlike the conventional manual operation of the gantry and the angle of view using a joystick or the like, the effect of greatly reducing the work load on the observer can be obtained.

According to the fifth aspect of the present invention, the region of interest is designated by, for example, a rectangle by the man-machine interface means, and the viewing direction is set so that the rectangular region is imaged at a predetermined image specific position. Is controlled, and the angle of view of the image input device is controlled according to the size and shape of the rectangular area, so that the object of interest can be automatically zoomed up and monitored. Instead of tracking a moving object that is not monitored, the observer specifies the target of interest in the image to the last, so that the observer can specify the object that the user really wants to pay attention to, and furthermore, a specific position in the image in which the attention area is set in advance In addition to automatically controlling the gantry so that images are captured at the same time, it also automatically zooms in on the area of interest, making it easier and more precise than conventional monitoring work. There is an advantage that it is possible to perform visual.

[Brief description of the drawings]

FIG. 1 is a block diagram of a monitoring support device according to a first embodiment of the present invention.

FIG. 2 is an explanatory diagram for describing a method of specifying a region of interest and an operation example thereof according to the first embodiment of the present invention.

FIG. 3 is a block diagram of a monitoring support device according to a second embodiment of the present invention.

FIG. 4 is an explanatory diagram for describing a method of specifying a region of interest and an operation example thereof according to a second embodiment of the present invention.

FIG. 5 is a block diagram of a monitoring support device according to a third embodiment of the present invention.

FIG. 6 is an explanatory diagram for describing a method of designating a region of interest and an operation example thereof according to Embodiment 3 of the present invention.

FIG. 7 is a block diagram of a conventional mobile object tracking device.

FIG. 8 is a flowchart for explaining the operation of a conventional mobile object tracking device.

FIG. 9 is an explanatory diagram showing an operation example of a conventional moving object tracking device.

[Explanation of symbols]

 Reference Signs List 101 image input means 102 man-machine interface means 103 combining means 104 image display means 105 area-of-interest designation means 106 gantry 107 gantry control means 301 image input means 302 man-machine interface means 303 combining means 304 image display means 305 area-of-interest designation means 306 moving direction detecting means 307 first image memory 308 second image memory 309 moving vector detecting means 310 position calculating means 311 switching means 312 gantry 313 gantry controlling means 501 image input means 502 man-machine interface means 503 combining means 504 image display means 505 attention area designation means 506 movement direction detection means 507 first image memory 508 second image memory 509 movement vector detection means 510 position calculation means 511 switch Step 512 Mount 513 Mount control 514 Angle of view control

Claims (5)

[Claims] 1. An image input means, a gantry for controlling a direction of a field of view of the image input means, a man-machine interface means for monitoring an image input by the image input means and designating a region of interest in the image, A monitoring support device comprising: a gantry control unit that controls a gantry based on positional information in an image of a region of interest specified by a machine interface unit. 2. The gantry control means controls the gantry so that a region of interest specified by the man-machine interface means is imaged at a predetermined specific position of an image obtained by the image input means after the gantry control. The monitoring support device according to claim 1, wherein: 3. An image input means, a gantry for controlling a direction of a field of view of the image input means, a man-machine interface means for monitoring an image input by the image input means and designating a region of interest in the image, A moving direction detecting means for detecting a moving direction of an object at a specific position of a preset image by using a plurality of images input by the input means, and controlling the gantry based on the region of interest and / or the moving direction And a gantry control means. 4. The gantry control means controls the gantry so that a region of interest specified by the man-machine interface means is imaged at a predetermined specific position of an image obtained by the image input means after the gantry control. 4. The monitoring support according to claim 3, wherein the gantry is controlled such that the region of interest is always imaged at a predetermined specific position of the image based on the moving direction detected by the moving direction detecting means. apparatus. 5. An image input unit having a zoom function,
A gantry for controlling the direction of view of the image input means, a man-machine interface means for monitoring the image input by the image input means and designating a region of interest in the image, and a plurality of images input by the image input means A moving direction detecting means for detecting a moving direction of an object at a specific position of a preset image, a gantry controlling means for controlling a gantry based on the region of interest and / or the moving direction, and a size of the region of interest. A monitoring support device comprising: an angle-of-view control unit that performs zoom control of an image input unit according to a shape and a shape.