JPH04126473A - Automatic tracking camera device - Google Patents

Abstract

PURPOSE:To allow the camera device to cope with the movement of plural objects by storing a characteristic relating to an object commanded by a command input means to a storage means, retrieving the object from a pattern picked up by an image pickup means based on the characteristic and tracking the object by a tracking means. CONSTITUTION:When a command input means 1 enters a command of tracking object, a tracking object storage means 5 stores picture information such as color distribution of the inside of a closed space being the tracking object and an external characteristic. Succeedingly, when a contour of the object stored by the means 5 is moved, a moving vector detection means 7 detects its moving vector by using an optional system. An arithmetic means 9 calculates a direction in which a video camera device 3 is moved so that the moving tracking object comes to the center of the pattern. Then a camera drive means 15 moves the direction of visual field picked up by the video camera device 3 in the direction calculated by the arithmetic means 9. Thus, the visual field is revised corresponding to the movement of the object and the camera device copes with the movement of plural objects.

Description

DETAILED DESCRIPTION OF THE INVENTION [Object of the Invention] (Industrial Application Field) The present invention relates to an automatic tracking camera device that automatically tracks and photographs a subject.

(Prior Art) In recent years, video cameras having a so-called autofocus function, which automatically focuses on a subject when photographing the subject, have become popular.

In a video camera having this autofocus function, a distance measurement field of view is fixedly set at the center of the photographing screen, and a subject within this distance measurement field of view is automatically focused. Therefore, when photographing, by capturing the subject within the distance measurement field of view, it is possible to automatically focus on the subject.

However, in this case, if the subject moves out of the distance measurement field of view, the object behind the subject will be in focus, and the subject will be out of focus, resulting in a blurred image of the subject. became.

Therefore, an automatic tracking device has been proposed that detects the movement of the subject and makes the size of the distance measurement field variable by manual setting, thereby preventing the above-mentioned shift in focus due to the movement of the subject (Japanese Patent Laid-Open No. 12177/1983). Public bulletin).

According to this conventional automatic tracking device, by manually changing the size of the tracking field of view and tracking based on the characteristics of the tracked subject, stable tracking operation can be achieved in response to changes in the size of the tracked subject. It can be carried out.

(Section 8 to be solved by the invention) However, in the conventional automatic tracking device described above, the tracked subject is tracked only within the screen being photographed,
By manually changing the size of the tracking field of view, the subject can be tracked over a wider range. Therefore,
If the tracked subject moves out of the screen, the tracked subject cannot be tracked, and by manually changing the size of the tracking field of view, the size of the tracked subject within the screen can be adjusted. It was expected that this would make it difficult to track the subject.

Furthermore, in the conventional automatic tracking device, no consideration was given to the tracking method when there are a plurality of objects to be tracked.

The present invention has been made in view of the above circumstances, and an object of the present invention is to provide an automatic tracking camera device that changes the field of view in response to the movement of a subject to be tracked, and is capable of responding to the movement of multiple subjects. shall be.

[Structure of the Invention] (Means for Solving the Problems) The automatic tracking camera device of the present invention includes a photographing means, an instruction input means for instructing a subject within a screen photographed by the photographing means, and an instruction input means. a storage means for storing characteristics related to the object specified by the storage means; a search means for searching for the object from within the screen photographed by the photographing means based on the characteristics stored in the storage means; and a search means using the retrieval means. The object of the present invention is to include a tracking means for tracking a subject to be photographed.

(Function) In the automatic tracking camera device of the present invention, the subject to be tracked is indicated within the screen shot by the shooting means. As a result, the subject to be tracked is specified, and the characteristics of the subject can be stored in the storage means.

Therefore, the search means searches for a subject based on the characteristics of the subject without being influenced by the size of the subject, and the tracking means automatically tracks the searched subject.

(Embodiment) Hereinafter, one embodiment of the present invention will be described in detail with reference to the drawings.

First, the configuration of the automatic tracking camera device of this embodiment will be explained with reference to the block diagram of FIG.

The instruction input means 1 is used to input an instruction for a subject to be tracked on a screen that is photographed by a video camera device 3 serving as a photographing means and displayed on the screen.

For example, instructions can be entered from outside using a keyboard, mouse, or touch panel, using a cursor among many objects displayed on the screen, or by pointing the object at the center of the screen. There is a method that allows you to specify the target to be tracked by tracing its outline.

In addition, for example, a computer or the like may input an instruction from the inside to automatically designate an object whose moving speed exceeds a predetermined value, or an object colored or arranged in a predetermined color as a tracking target.

The tracking object storage means 5 stores information such as a human face, a soccer ball, etc.
This is a memory element/stage that stores image information such as the color distribution inside a closed space (for example, the outline) of the object specified by the instruction input means 1, external shape characteristics, etc.

The motion vector detection means 7 detects a motion vector when the outline (internal color distribution information) of the object stored in the tracking object storage means 5 moves. Furthermore, this motion vector detection means 7 can calculate a motion vector even if the shape of the object's outline changes to some extent as long as the internal color distribution, for example, information such as green next to red, does not change. It is possible to detect movement in a closed space even if the color information inside the outline changes so that the object is facing backwards.

Methods for detecting such movements include ■) Method 2: Shifting two consecutive images little by little to find the point where the difference between the images is the minimum, and calculating the amount of movement from the deviation of this minimum position. ) A method of calculating the cross-correlation function of two images and calculating the amount of movement from the deviation that maximizes this cross-correlation function 3) A method of calculating the amount of movement by the ratio between Fourier transform images 4) A method of calculating the amount of movement by calculating the amount of movement from the ratio between the Fourier transform images 4) At a certain position of the image A method of determining the amount of movement from the relationship between the spatial gradient and the difference between images is known.

This motion vector detection means 7 always detects where the tracking target is located on the screen and how fast it is moving.
Each tracked target has information about the size of the closed space of the tracked target on the screen.

The calculation means 9 stores information on the direction and distance in which the camera is moved, and the time at which the camera was moved, and when the tracking target moves, the video camera moves so that the tracking target comes to the center of the screen. The amount of movement when changing the visual field direction of the device 3 and the amount of driving when adjusting the size of the visual field are calculated.

The field of view adjustment means 11 adjusts the size of the field of view of the video camera device 3.

When the tracking target is lost on the screen, the tracking target search means 13 forcibly widens the field of view using the field of view adjustment means 11, rotates the video camera device 3 in the direction in which the target disappears, and gives instructions to move the field of view. The camera driving means 15 is provided with the following information. This movement in the visual field direction may be performed simultaneously with the expansion of the visual field, or only one of them may be performed.

When the tracking target appears again on the screen after the tracking target has been lost on the screen, the tracking target recognition means 13a compares the tracking target with the target stored in the tracking target storage means 5, and they match. It is something that recognizes that.

The camera drive means 15 changes the viewing direction according to the amount of movement calculated by the calculation means 9.

Next, the automatic tracking operation in the automatic tracking video camera device of this embodiment will be explained.

First, when an instruction for a tracking target is input using the instruction input means 1, the tracking target storage means 5 stores image information such as the color distribution inside a closed space (for example, an outline) and external shape characteristics of the tracking target. do.

Subsequently, when the contour (internal color distribution information) of the object stored in the tracking object storage means 5 moves, the motion vector detection means 7 detects the motion vector using any of the above-mentioned methods.

The calculation means 9 calculates the direction in which the video camera device 3 should move so that the moving tracking target comes to the center of the screen. When the speed of moving the video camera device 3 is fast, or when the distance to move the video camera device 3 is long, the calculation means 9 gives an instruction from the calculation means 3 to the field of view adjustment means 11 to widen the field of view.

Conversely, if the distance to which the video camera device 3 is moved is short and the size of the outline stored in the tracking target storage means 5 is smaller than the initially stored outline, the tracking target on the screen may be The field of view is narrowed by the field of view adjusting means 11 until the size of the memorized outline is reached.

Subsequently, the camera drive means 15 moves the viewing direction for photographing with the video camera device 3 in the direction calculated by the calculation means 9. In addition, at this time, the tracking object storage means 5 has an area for storing a plurality of objects, and classifies the plurality of tracking objects by codes, for example, numbers including information related to priority or search order, or information related to characteristics of the objects. Based on this stored information, if there are multiple objects to be tracked, the computing means 9 calculates the moving distance of the video camera W3 so as to track the center of gravity of each of the multiple objects. and the direction are calculated, and the viewing direction and the like are set.

For example, as shown in Figure 2, there are three tracking targets, and the third
When each tracking target moves in the direction of the arrow shown in the figure (Xo = X+, Xo = Y+, ZO→X1),
The direction of the video camera device 3 is moved so that the centers of gravity G of the three objects shown in FIG. 4 come to the center C of the screen (see FIG. 5).

Here, the center of gravity G shown in FIGS. 4 and 5 is an example obtained by giving equal weight to (X, Y, Z) of each tracking target.

In FIGS. 6 and 7, each tracking target x,,y,.

This is an example in which the reciprocal of the priority number corresponding to Zl is calculated as the weight of each tracking target.

The focus adjustment means 17 is a means for automatically adjusting the focus. As in this embodiment, when there are multiple objects to be tracked, focus is placed on the object with the highest priority (highest number) associated with the object.

The focus adjustment means 17 obtains information on the position of the object to be focused on on the screen from the performance means 9, and also obtains information on the position of the single and plural tracking objects on the screen from the visual field adjustment means 11 or the motion vector detection means 7. , and in what direction and at what speed, the system focuses on the object to be tracked and adjusts the size of the field of view so as not to lose the object on the screen.

Further, when the tracking target is lost on the screen, the tracking search unit 13 forcibly widens the field of view using the visual field adjusting unit 11, and instructs the calculating unit 9 to rotate the video camera device 3 in the direction in which the target disappears. give.

Furthermore, it includes tracking object recognition means 13a for recognizing that when the tracking object appears on the screen again, it matches the object stored in the tracking object storage means 5. This tracking object recognition means 13a focuses on a moving closed space on the screen and finds an object whose color distribution matches or is very similar to the color distribution information stored in the tracking object storage means 5. .

As described above, according to the automatic tracking camera device of this embodiment, as an arbitrary object moves, the video camera device can automatically move the object so that the object is in the center of the screen to track and photograph the object. I can do it.

Furthermore, even if the shape of the memorized tracking target changes to some extent, it can be recognized that it is the same as the originally memorized target based on image information such as color distribution inside a closed space (for example, an outline) and external shape characteristics. Therefore, even if there is a single or multiple tracking target, the field of view can be automatically adjusted so that the tracked target appears on the screen in a natural way, and even if the tracked target is lost. When you enlarge your field of view,
Alternatively, the tracking target can be searched by rotating the camera.

That is, when the object appears on the screen again, it is possible to recognize that it is the object that was first memorized and to resume tracking. Further, when there are multiple objects to be tracked, the objects can be stored together with the priority to be tracked, and the camera is configured to focus on the one with the higher priority. Note that by manual switching, it is also possible for a person to manually take pictures as in a conventional video camera device.

[Effects of the Invention] As explained above, according to the present invention, the object is searched from within the screen based on the characteristics of the object.
The field of view can be changed in response to the movement of a subject, and the effects can be achieved such as being able to cope with the movement of a plurality of subjects.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of the present invention, FIG. 2 is a diagram showing a case where there are three objects to be tracked, and FIG. 3 is a diagram showing the situation when the three objects move. Figures 4 and 5 are diagrams showing the case where the video camera device is moved so that the center of gravity of the tracked target after moving is at the center of the screen, and Figures 6 and 7 show the center of gravity of multiple tracked targets. This figure shows how to find the center of gravity by weighting it with the reciprocal of the priority of each object. 1...-Instruction input means 39. Video camera device 5...Tracked object storage means 7...Motion vector detection means 9...Calculation means 11...Visual field adjustment means 13...Tracked object search means 13a...Tracked object recognition means 15.・Camera driving means

Claims (4)

[Claims] (1) a photographing means; an instruction input means for specifying a subject within the screen photographed by the photographing means; a storage means for storing characteristics of the subject specified by the instruction input means; and storage in the storage means. An automatic tracking camera device comprising: a search means for searching for a subject from within a screen photographed by the photographing means; and a tracking means for tracking the subject searched for by the search means, based on the characteristics of the camera. . (2) The search means is characterized by having a field of view adjustment means for adjusting the field of view direction and/or the size of the field of view of the photographing means when the object cannot be searched within the screen photographed by the photographing means. The automatic tracking camera device according to claim 1. (3) The automatic tracking camera device according to claim 1 or 2, wherein when there are a plurality of subjects to be photographed by the photographing means, the storage means stores the characteristics of each subject along with a code. (4) The automatic tracking according to claim 3, wherein the code assigned by the storage means is a coefficient related to a search order, and the subject to be photographed by the photographing means is focused in accordance with this search order. camera equipment.