JPH06347265A - Monitoring apparatus of moving object - Google Patents

Abstract

PURPOSE:To automatically track and monitor a moving object over a wide space range by using photographing apparatuses whose number is as small as possible in an apparatus which tracks and monitors the moving object. CONSTITUTION:Two fixed photographing apparatuses 11A, 11B respectively photograph the whole of a space region to be monitored. One movable photographing apparatus 12A photographs only a moving object inside the region to be monitored, and its focal distance and its photographing direction are variable. Moving-object detection parts 20A, 20B respectively detect the position of the moving object from the inside of images from the fixed photographic apparatuses 11A, 11B. A moving-object recognition part 60 decides the three-dimensional position of the moving object by a binocular stereoscopic vision technique from the position of the moving object which has been detected, and it estimates the future three-dimensional position of the moving object. An input-apparatus control part 70 and a driving device 50 variably control the photographing direction and the focal distance of the movable photographing apparatus 12A on the basis of the estimated future position of the moving object.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a moving object monitoring apparatus which is suitable for application to a crime prevention system, a traffic monitoring system, a target tracking system and the like and which captures an image while tracking a moving object.

[0002]

2. Description of the Related Art FIG. 1 shows the configuration of a conventional moving object monitoring apparatus.

In this conventional apparatus, generally, a video input unit 1 including one or a plurality of photographing devices, a moving object detection unit 2 for detecting a moving object in the input video, and a detected movement. The result display unit 3 for displaying the object on the monitor and the result recording unit 4 for recording the detected moving object are configured.

Here, the photographing device of the image input section 1 is a movable type whose focal length and photographing direction can be changed, or a fixed type. For the movable type,
The driving device 5 operated by the user 6 changes the focal length and the shooting direction.

In this structure, the image input from the image input unit 1 is processed by the moving object detection unit 2 to check whether or not a moving object exists, and only the image in which the moving object is detected is displayed in the result display unit. 3 and is stored in the result recording unit 4.

Then, the user 6 adjusts the focal length and the photographing direction of the photographing device as necessary so that the moving object can be continuously observed by looking at the image on the result display section 3.

[0007]

One problem with this conventional monitoring device is that when monitoring is performed over a wide spatial range,
That is, a huge number of image capturing devices are required.

That is, when a fixed type image pickup device is used, it is necessary to make the focal length sufficiently long in order to project a moving object in a size sufficient for monitoring. This is because when the image is taken with a short focal length, the moving object appears to be very small, and even if it is enlarged and displayed by image processing, sufficient resolution may not be obtained. However, if you use an imaging device with a long focal length, the field of view becomes narrower,
The moving object deviates from the visual field range in a short time. Therefore, in order to cover a wide monitoring range, it is necessary to prepare a large number of video devices and arrange them so that their respective visual field ranges partially overlap.

On the other hand, when the movable type photographing apparatus is used, the above problem does not occur. However, since it is necessary for the user to manually adjust the focal length and the shooting direction, it is a very difficult and painstaking task for the user to follow a moving object in real time.

Therefore, an object of the present invention is to enable tracking and monitoring of a moving object over a wide space range with only a relatively small number of photographing devices.

[0011]

A moving object monitoring apparatus of the present invention comprises a first photographing means for photographing the entire monitoring area, and a focal length and photographing for photographing a moving object in the monitoring area. An object position determining unit that determines a three-dimensional position of a moving object based on an image from the second image capturing unit whose direction is variable and the first image capturing unit, and the second image capturing unit keeps the moving object in the visual field at all times. A control means for variably controlling at least one of the focal length and the photographing direction of the second photographing means based on the three-dimensional position from the object position determining means so that the object can be captured and photographed in a desired size. And

In a preferred embodiment, at least two fixed-type photographing devices installed at different places are used as the first photographing means. These fixed imaging devices are set to have short focal lengths so that the entire surveillance area is within the field of view. Further, at least one movable type photographing device is used as the second photographing means. This movable photographing device is provided with a drive device for variably controlling the focal length and the photographing direction so that a moving object in the monitoring area can be photographed in a large size suitable for monitoring.

[0013]

According to the present invention, a first photographing means having a short focal length for monitoring a wide area in which a moving object moves,
A moving object is automatically tracked and monitored mainly by cooperatively using a second imaging means having a variable focal length and an imaging direction for monitoring only a moving object.

First, the three-dimensional position of the moving object shown in the video is determined based on the video for wide-area monitoring from the first photographing means. Next, the photographing direction and the focal length of the second photographing means are variably controlled based on the three-dimensional position. Thereby, the second image capturing means captures the moving object in a sufficiently large size while always capturing the moving object in the visual field.

[0015]

FIG. 2 is a block diagram showing the configuration of a preferred position embodiment of the moving object monitoring apparatus of the present invention.

As shown in FIG. 2, the moving object monitoring apparatus has a fixed photographing device 11 having a fixed focal length and fixed photographing direction.
A video input unit 10 including A and 11B and a movable photographing device 12A whose focal length and photographing direction are variable, and a fixed photographing device 11
Moving object detection units 20A and 20B that detect a moving object in the images captured by A and 11B, and a moving object recognition unit 60 that predicts the future three-dimensional position of the detected moving object.
An input device control unit 70 that generates a drive command based on the predicted future position of the moving object, a drive device 50 that changes the focal length and the shooting direction of the movable imaging device 12A according to the drive command, and the movable imaging device. The result display unit 30 that visually displays the image captured by the image capturing device 12A and the result recording unit 40 that stores the image are provided.

The video input unit 10 includes two fixed photographing devices 1
1A, 11B and one movable imaging device 12A are included.
The three imaging devices 11A, 11B, and 12A, as a set, are in charge of imaging one spatial region to be monitored (hereinafter referred to as a monitoring region).

If the monitoring area is very wide and one set of imaging devices 11A, 11B, 12A is not sufficient to cover it, one or more additional imaging devices can be added. Further, in order to monitor the monitoring area in more detail, a larger number of fixed photographing devices and movable photographing devices can be combined into one set.

The above-mentioned two fixed photographing devices 11A and 11
B is installed at a different position, the imaging directions are fixed parallel to each other toward the surveillance area, and the focal length is fixed to a sufficiently short focal length so that the entire surveillance area can be included in the visual field. . These fixed photographing devices 11A, 11
B is for photographing the entire surveillance area.
The image captured by this is called a wide area monitoring image.

On the other hand, the movable photographing device 12A has a variable photographing direction and focal length, the variable range of the photographing direction includes all positions in the monitoring area, and the variable range of the focal length is fixed photographing devices 11A and 11B. The focal length range is longer than the focal length of. The driving device 50 controls the shooting direction and the focal length of the movable shooting device 12A. The movable image capturing device 12A is for capturing only a moving object within the monitoring area, and hereinafter, the image captured by this is referred to as a moving object monitoring image.

These three photographing devices 11A, 11B and 1
2A outputs one frame of video imaged in each cycle at a constant cycle. For example, in the current mainstream video cameras, the period is 1/30 second.

In this embodiment, two fixed photographing devices 11 are used.
The position of the moving object is determined from the wide area monitoring images from A and 11B, and the movable imaging device 12B is moved according to this information to perform tracking and imaging of the moving object. Figure 3
Shows the operation flow of this embodiment. Below, FIG.
The configuration and operation of this embodiment will be described in detail with reference to FIG.

The two wide-area monitoring images output from the fixed photographing devices 11A and 11B are input to the moving object detection units 20A and 20B provided for each fixed photographing device (S1). Each moving object detection unit 20A, 20B temporarily accumulates each frame of the image input in each cycle in an internal storage device, and calculates the difference between the frame input in the latest cycle and the frame input in the previous cycle. Calculate Then, the area in the frame where the difference value is a significant value is recognized as an area in which the moving object exists, and the moving object recognition unit 60 recognizes information indicating the position of the moving object existing area in the frame.
(S2).

The moving object recognizing unit 60 includes position information of the moving object existing area input from the two moving object detecting units 20A and 20B, and two predetermined fixed photographing equipments 11.
The three-dimensional position of the moving object in the monitoring area is calculated by the binocular stereoscopic (or stereoscopic) method from the three types of information of the installation positions of A and 11B and the focal length (S3).

Here, the binocular stereoscopic technique is a technique for calculating the three-dimensional position of an object from a plurality of images by applying the principle of triangulation, and the details thereof are described in, for example, the book "Computer Vision" Masahiko Taniuchida. Written in 1990, Maruzen Co., Ltd., page 119.

Next, the moving object recognition unit 60 accumulates the three-dimensional position information of the moving object in each cycle in an internal storage device,
Then, based on the position information, the moving direction and moving speed of the moving object are calculated and stored in the internal storage device (S).
4).

Further, the moving object recognition unit 60 predicts the three-dimensional position of the moving object at a future time point after a predetermined time from the present time point by using the Kalman filter from the accumulated time series data of the moving direction and the moving speed, and Information on the predicted future position is given to the input device control unit 70 (S5).

Note that, as the future time point to be the target of this prediction, in consideration of the shooting direction and the tracking capability of the focal length of the movable shooting device 12A driven by the drive control device 50, they are set at the predicted future position. The future is long enough to be sure to follow, and the future is close enough that the predicted position does not substantially differ from the actual destination of the moving object. A time point later is selected.

The input device control unit 70 receives the predicted position information, and a moving object that will be present at the predicted position at a future time is of a sufficiently large size suitable for monitoring in the visual field of the movable photographing device 12A. As described in step S6, a drive command for changing one or both of the shooting direction and the focal length of the movable shooting device 12A is generated and given to the drive device 50 (S6).

The drive unit 50 follows the drive command,
The shooting direction and the focal length of the movable shooting device 12A are controlled (S7). As a result, the movable image capturing device 12A always captures a moving object in the field of view and captures it with a sufficiently large size suitable for monitoring.

The moving object monitoring image output from the movable photographing device 12A is used as a result display section 30 and a result recording section 4.
It is input to 0 (S8). The result display unit 30 is, for example, CR
The result recording unit 40 is a T display, and the result recording unit 40 is a large-capacity storage device suitable for storing video information, such as a disk storage device. These result display unit 30 and result recording unit 40
Respectively, the whole frame of the moving object monitoring video, or a partial video in it for the purpose of monitoring, the shooting time,
It is displayed and stored together with accompanying information such as the number of the movable photographing device and the three-dimensional position of the moving object (S9).

The preferred embodiment of the present invention has been described above. However, the present invention is not limited to this embodiment and can be carried out in various different modes without departing from the scope of the invention. It is a thing. For example, in the above embodiment, the three-dimensional position of the moving object was calculated by the binocular stereoscopic method based on the wide area monitoring images from the two image capturing devices. However, when the actual size of the moving object is constant, Is 1
It is also possible to determine the three-dimensional position of the moving object based on the two-dimensional position and size of the moving object in the wide-area monitoring video imaged by the single imaging device.

[0033]

As described above, according to the present invention,
Since the three-dimensional position of the moving object is determined from the wide area monitoring image and the shooting direction and the focal length of the shooting means for shooting the object are variably controlled based on this, the moving object is moved using a small number of shooting devices. It is possible to automatically track and monitor an object.

[Brief description of drawings]

FIG. 1 is a block diagram showing a configuration of a conventional device.

FIG. 2 is a block diagram showing the configuration of a position embodiment of the present invention.

FIG. 3 is a flowchart showing the operation of the embodiment.

[Explanation of symbols]

 10 Image Input Section 11 Fixed Shooting Equipment 12 Movable Shooting Equipment 20 Moving Object Detection Section 30 Result Display Section 40 Result Recording Section 50 Drive Device 60 Moving Object Recognition Section 70 Input Equipment Control Section

Claims (5)

[Claims] 1. An apparatus for photographing a moving object in a surveillance area, comprising: a first photographing means for photographing the whole of the surveillance area; and a focal length for photographing the moving object in the surveillance area. And a second photographing means having a variable photographing direction, an object position deciding means for deciding a three-dimensional position of the moving object based on an image from the first photographing means, and a second photographing means for moving the object. At least one of the focal length and the photographing direction of the second photographing means is variably controlled based on the three-dimensional position from the object position determining means so that the object can always be captured in the visual field and photographed in a desired size. A moving object monitoring device comprising: a control unit. 2. The apparatus according to claim 1, wherein the first imaging unit includes a plurality of imaging devices installed at different positions for imaging the entire monitoring area, and the object position determination is performed. The means includes means for calculating a three-dimensional position of the moving object by a binocular stereoscopic technique based on the two-dimensional positions of the moving object in a plurality of images from the plurality of photographing devices. A moving object monitoring device. 3. The apparatus according to claim 1, wherein the first image capturing unit includes one image capturing device that captures an image of the entire monitoring area, and the object position determining unit captures the one image capturing unit. A moving object monitoring device comprising means for calculating a three-dimensional position of the moving object based on a two-dimensional position and a size of the moving object in an image from a device. 4. The apparatus according to claim 1, wherein the control unit calculates a moving direction and a moving speed of the moving object based on the three-dimensional position of the moving object determined at different times. Means for predicting a future three-dimensional position of the moving object based on a moving direction and a moving speed of the moving object; and the second based on the predicted future three-dimensional position.
And a means for changing at least one of the focal length and the photographing direction of the photographing means. 5. A first photographing means for photographing the entire surveillance area, and a second photographing means for photographing a moving object in the surveillance area, the photographing direction and the focal length of which are variably controllable. In the control device used in the moving object monitoring apparatus, the object position determining means for determining the three-dimensional position of the moving object based on the image from the first photographing means, and the second photographing means for moving the moving object. At least one of the focal length and the photographing direction of the second photographing means is variably controlled based on the three-dimensional position from the object position determining means so that the object can always be captured in the visual field and photographed in a desired size. And a control signal generating means for generating a signal for the control of the moving object monitoring device.