KR20040031968A - Unidentified People Tracking Device Using Dual Cameras And Method Thereof - Google Patents

Abstract

PURPOSE: An apparatus for tracking a trespasser using a double camera and a method thereof are provided to track a trespasser while taking photographs of a whole security area. CONSTITUTION: A first camera(10) is fixed by one end for taking photographs in an axial direction. An fisheye lens(12) is installed at the first camera for mapping images of a peripheral subject of the first camera as one image. A first rotating part(14) is installed to be rotative on the axis of the first camera. A second camera(18) is mounted on a second rotating part(16) rotatively movable at a predetermined angle independently from the rotary movement of the first rotating part. A judging unit judges the existence of a trespasser from a plurality of image signals of the first camera. A calculating unit calculates coordinates data of the trespasser based on the image signals of the first camera according to the judgement. A driving signal output unit generates driving signals to be transmitted to the first and second rotating parts based on the coordinates data to take a photograph of the trespasser. An image storage unit stores the images of the first camera and the second camera.

Description

Intruder tracking device using dual cameras and its method {Unidentified People Tracking Device Using Dual Cameras And Method Thereof}

The present invention relates to a security surveillance camera device, and more particularly, to an intruder tracking device and method using a double camera that the first camera having a fisheye lens determines whether the intruder invaded and tracks the second camera intruder. It is about.

In general, a security or unmanned security system has one CCD camera, a left and right drive unit for rotating the CCD camera left and right, and a turning drive unit for turning up and down. In addition, in the case of a camera installed in a large area, a zoom lens is installed and has a zoom function.

Therefore, such a conventional CCD camera takes a wide area at a constant angular velocity while taking a round trip, and a security officer monitors and records it at the center.

However, such a conventional unmanned camera device has a disadvantage in that it is impossible to monitor at all an area outside the shooting range of the camera. Therefore, in order to prevent this, a method of minimizing blind spots of surveillance has been used by installing a plurality of cameras in various places and adjusting the rotation speed. However, this method also has a limitation that it is not possible to always monitor and photograph the entire interior at the same time.

In particular, when an intruder is captured by the camera, a zoom function is used to enlarge the image. At this time, if the intruder moves several steps, the intruder's image disappears from the screen.

Therefore, there has been a demand for the development of a system that always photographs the entire surveillance area, and automatically tracks the intruder's trajectory when the intruder invades and places the intruder in the center of the camera image.

Therefore, the present invention has been made in view of the above problems, the first object of the present invention, an intruder using a dual camera that can always shoot, record and monitor the entire indoor or surveillance area (security guard area). It is to provide a tracking device and a method thereof.

It is a second object of the present invention to provide an intruder tracking device and method using a dual camera that can automatically detect the moving position of an intruder when the intruder occurs in the security target area and take a picture of the intruder while the camera tracks the intruder. It is.

In addition, it is possible to provide an intruder tracking device using the dual camera and its method that can be easily installed and used by anyone by being able to automatically track the location of the intruder, but can be manufactured using a simple and low-cost equipment.

An object of the present invention as described above, the first camera 10, one end is fixed to shoot in the axial direction; Optical means installed in the first camera 10 to map an image of a subject around the first camera 10 into a single image; A first rotating part (14) rotatably installed on an axis of the first camera (10); A second camera (18) mounted on the first rotating portion (14) and mounted on a second rotating portion (16) rotatable by a predetermined angle independently of the rotational movement of the first rotating portion (14); Determination means for determining the presence or absence of the intruder (50) from the plurality of image signals of the first camera (10) through the optical means; Calculating means for calculating coordinate data of the intruder 50 based on the video signal of the first camera 10 when a command indicating that the intruder 50 has invaded by the determining means is output; In order for the second camera 18 to photograph the intruder 50, a driving signal of the first rotating part 14 and the second rotating part 16 is generated based on the coordinate data, so that the corresponding rotating parts 14 and 16 are generated. A driving signal output unit 36 for transmitting to each; And it can be achieved by the intruder tracking device using a dual camera, characterized in that it comprises an image storage means 40 for storing the images of the first camera 10 and the second camera 18.

In addition, computing means for comparing and calculating a plurality of video signals of the second camera (18); And a tracking vector calculating means for calculating a tracking vector representing a moving direction of the intruder 50 from the calculating means, wherein the driving signal output unit 36 is configured to perform the first rotation unit 14 based on the tracking vector. ) And the driving signals of the second rotating part 16 are preferably transmitted to the corresponding rotating parts 14 and 16, respectively.

The image storage means 40 is more preferably a VCR, a DVR, a hard disk, a CD-RW, a DVD-RAM, a magnetic recording tape, a memory stick, or the like.

In addition, the first camera 10 is most preferably fixed to the ceiling 20, and installed so as to shoot vertically downward.

In addition, the first rotating part 14 may be rotatable within a range of 0 ° to 360 °, and the second rotating part 16 may be rotatable within a range of 0 ° to 90 °.

The coordinate data may include the distance R from the center of the image to the intruder 50 and the counterclockwise rotation angle Θ from the reference axis (in this embodiment, the X axis of FIG. 7). desirable.

The method may further include selectively transmitting the stored video signal or the first and second camera video signals through the Internet.

Preferably, the optical means is a fisheye lens 12.

Alternatively, the optical means may include: a column 94 having one end fixed and the other end extending along an axis of the first rotating part 14 to protrude out of the first rotating part 14; A convex mirror 90 having one surface fixed to the column 94 and the other surface forming a mirror; A first camera 10 facing the convex mirror 90 and spaced apart from the convex mirror 90 by a predetermined distance; And a transparent case 92 on which one surface is fixed to the column 94, the convex mirror 90 is installed inside, and the first camera 10 is fixed to one end thereof.

Alternatively, the optical means may include an annular convex mirror 70 attached to the first camera 10 to reflect an image of a subject around the first camera 10; A concave mirror 72 installed to be spaced apart from the annular convex mirror 70 by a predetermined interval to reflect the image reflected by the annular convex mirror 70 to the first camera 10; And a transparent case 74 for supporting the annular convex mirror 70 and the concave mirror 72.

The object of the present invention as described above is also obtained from the plurality of image signals of the first camera 10 through the assembly of the fisheye lens 12, the convex mirror 90 or the annular convex mirror 70 and the concave mirror 72. Determining the presence or absence of the intruder 50; Calculating coordinate data of the intruder 50 based on an image signal of the first camera 10 when a command indicating that the intruder 50 has invaded is output; In order for the second camera 18 on the second rotating part 16 mounted on the first rotating part 14 to photograph the intruder 50, the first rotating part 14 and the second rotating part based on the coordinate data. Generating driving signals of (16) and transmitting them to the corresponding rotating units (14, 16), respectively; Calculating a tracking vector representing a moving direction of the intruder by comparing a plurality of image signals of the second camera 18; And generating driving signals of the first rotating part 14 and the second rotating part 16 based on the tracking vector and transmitting the generated driving signals to the corresponding rotating parts 14 and 16, respectively. Of course, it can also be achieved by the intruder tracking method used.

The method may further include storing the video signal of the first camera 10 and the video signal of the second camera 16, respectively.

Other objects, specific advantages and novel features of the present invention will become more apparent from the following detailed description and the preferred embodiments associated with the accompanying drawings.

1A is a schematic side view of a dual camera device 1 equipped with a fisheye lens 12 according to a first embodiment of the present invention;

1B is a schematic side view of a dual camera apparatus 1 equipped with a convex mirror 90 according to a second embodiment of the present invention;

1C is a schematic side view of a dual camera apparatus 1 equipped with an annular convex mirror 70 and a concave mirror 72 according to a third embodiment of the present invention;

2 is a plan view of the dual camera device 1 shown in FIG. 1A,

3 is an installation state diagram of an intruder tracking device using a dual camera according to the present invention,

4 is an internal block diagram of the tracking driver 30 of FIG.

5 is an example of an image in which the first camera 10 according to the present invention photographs the room through the fisheye lens 12;

6 is an example of the image when the intruder 50 intrudes in the image shown in FIG. 5,

FIG. 7 is an example of an image defining and calculating coordinates of an intruder 50 in the image shown in FIG. 6.

8 is an example of an image taken by the second camera 18 of the present invention moved according to the coordinates calculated in FIG.

9 is an example of an image for determining a moving direction of an intruder 50 by comparing and calculating a plurality of images photographed by a second camera;

FIG. 10 is an example of an image for calculating the tracking vector 64 from the moving direction of FIG. 9.

<Description of the symbols for the main parts of the drawings>

1: dual camera device, 10: first camera,

12: fisheye lens, 14: first rotating part,

16: second rotating part, 18: second camera,

20: ceiling, 22: base,

30: tracking driver, 40: DVR,

50: intruder, 52: door,

54: indoor, 60: second camera image,

64: tracking vector, 70: annular convex mirror,

72: concave mirror, 74, 92: transparent case,

90: convex mirror, 94: column.

Hereinafter, with reference to the accompanying drawings with respect to the configuration of the intruder tracking device and method using a dual camera according to the present invention will be described in detail.

FIG. 1A is a schematic side view of a dual camera apparatus 1 in which a fisheye lens 12 is installed according to a first embodiment of the present invention, and FIG. 2 is a plan view of the dual camera apparatus 1 shown in FIG. 1A. As shown in FIGS. 1A and 2, a circular base 22 is fixed to the ceiling 20 to mechanically fix the dual camera apparatus 1.

The first camera 10 is fixed vertically downward in the center of the base 22. An example of such a first camera 10 is a low-cost CCD camera, a web camera, and the like. In addition, a fisheye lens 12 is attached to the lens region of the first camera 10. The fisheye lens 12 is configured to distort (or map) an indoor image so as to enter a screen.

The first rotating part 14 has an annular shape as a whole, and is provided with a servo motor (not shown) and a deceleration gear train (not shown) therein. The first rotating part 14 is installed at the base 22 and has a circumference of the first camera 10. It is configured to be able to rotate within the range of ° ~ 360 °. Therefore, the first rotating unit 14 may rotate the circumference of the first camera 10 by a predetermined angle (for example, 10 ° clockwise) according to a driving signal received from the outside.

The second rotating part 16 is mounted on the first rotating part 14, and is configured to be able to rotate within the range of about 0 ° to 90 ° in the vertical direction independently of the rotational movement of the first rotating part 14. . For this purpose, a servo motor (not shown) and a deceleration gear train (not shown) are provided therein. Therefore, the second rotating unit 16 may rotate by a predetermined angle (for example, 10 ° in the upward direction) according to the driving signal received from the outside.

1B is a schematic side view of a dual camera apparatus 1 equipped with a convex mirror 90 according to a second embodiment of the present invention. As shown in FIG. 1B, one end of the column 94 is fixed to the base 20 and extends along the axis of the first rotating part 14, but is independent of the rotational movement of the first rotating part 14. It is configured to be spaced apart from each other.

The transparent case 92 is formed of a transparent material such as acrylic, glass, or synthetic resin, and one end is fixed to the column 94. A convex mirror 90 is installed therein, and this convex mirror 90 is also fixed to the other end of the column 94 at the same time. The first camera 10 is located on the axis of the column 94 and faces the convex mirror 90 and is spaced apart from the convex mirror 90 by a predetermined distance. The first camera 10 is fixed in position by the transparent case 92. Accordingly, the image of the subject around the first camera 10 is reflected by the convex mirror 90 through the transparent case 92 and then incident to the first camera 10. Wiring (not shown) of the first camera 10 is arranged along the inside of the transparent case 92 and the column 94 to reach the base 20.

1C is a schematic side view of a dual camera apparatus 1 equipped with an annular convex mirror 70 and a concave mirror 72 according to a third embodiment of the present invention. As shown in FIG. 1C, a hemispherical transparent case 74 is installed at the front of the first camera 10, and an annular convex mirror 70 and a concave mirror 72 are disposed inside the transparent case 74. do.

The annular convex mirror 70 is located on the axis of the first camera 10, and the outer diameter portion forms the convex mirror. The concave mirror 72 is also located on the axis of the first camera 10 and is spaced apart from the annular convex mirror 70 by a predetermined distance. Accordingly, the image of the surrounding subject of the first camera 10 is first reflected by the annular convex mirror 70 through the transparent case 74 and then secondly reflected by the concave mirror 72 to be the first camera 10. Incident).

3 is an installation state diagram of an intruder tracking device using a dual camera according to the present invention. As shown in FIG. 3, the dual camera apparatus 1 is fixed to the ceiling 20 of the interior 54, and the door 52 and the intruder 50 are located in the interior 54. In addition, the output signal of the dual camera apparatus 1 is transmitted to the tracking driver 30, the output image of the tracking driver 30 is transmitted to the VCR or DVR (Digital Video Recorder) device 40 is recorded. The internal structure of the tracking driver 30 will be described in detail with reference to FIG. 4.

The video signal of the camera is transmitted from the dual camera device 1 to the tracking driver 30, and the camera driving signal, power, and camera operation signal are transmitted from the tracking driver 30 to the dual camera device 1.

4 is an internal block diagram of the tracking driver 30 of FIG. 3. As shown in FIG. 4, the control unit 32, the image signal input unit 31, the image signal output unit 33, the memory 35, the power supply unit 34, and the driving signal are largely located inside the tracking driver 30. An output unit 36 is included.

The video signal input unit 31 is a port to which the video signal of the first camera 10 and the video signal of the second camera 16 are input. The video data of the video signal input unit 31 is transmitted to the controller 32.

The controller 32 may be implemented by a microprocessor, a microcomputer, a CPU, or the like. The controller 32 analyzes each image data for each frame and determines and processes the image data according to a predetermined program.

The video signal output unit 33 transmits and records an image to a recording apparatus such as a VCR or a DVR 40. The power supply unit 34 has a configuration for supplying power required for the drive signal output unit 36 and the dual camera apparatus 1. The memory 35 is a storage place for temporarily storing, comparing, and processing images by frame. The drive signal output unit 36 has a configuration for generating and outputting a control signal for independently driving the first and second rotary units 14 and 16, respectively. When the position coordinate to be moved from the control unit 32 is inputted, it is converted into moving angle data of each rotating unit, and converted into a predetermined electric signal and transmitted to each rotating unit 14 and 16. To this end, the first drive signal output unit 36a and the second drive signal output unit 36b are provided therein.

In addition, if necessary, in order to receive image data from the outside, and to control the dual camera device 1 from the outside, a LAN card (not shown) may be installed inside the tracking driver 30 and the Internet (not shown) may be connected. have.

Hereinafter, a tracking method of an intruder tracking device using a dual camera according to the present invention having the above configuration will be described in detail with reference to the accompanying drawings.

FIG. 5 is an example of an image in which the first camera 10 according to the present invention photographs a room through the fisheye lens 12. Since the first camera 10 and the fisheye lens 12 are fixed, the image of the room 54 is distorted by the fisheye lens 12 and enters the camera lens. That is, the door 52 of FIG. 3 is photographed by the crushed door 52 in FIG. That is, in the dual camera apparatus 1 according to the present invention, in the absence of the intruder 50, only the first camera 10 is operated and the second camera 18, the first and second rotating parts 14 and 16 are not operated. Do not.

FIG. 6 is an example of an image when an intruder 50 intrudes in the image shown in FIG. 5. As shown in FIG. 6, the controller 32 compares the image of the first camera 10 for each frame (FIGS. 5 and 6). When the change occurs in the image data, the intruder 50 intrudes. I think that. This is because the data value of the pixel of the area occupied by the intruder 50 in FIG. 6 is different from the pixel data value of the corresponding area in FIG. 5.

FIG. 7 is an example of an image for defining and calculating the coordinates of the intruder 50 in the image shown in FIG. 6. As shown in FIG. 7, the controller 32 calculates coordinates from the center of the camera image to the center of the image of the intruder 50. That is, the radial radius R to the center of the image of the intruder 50 and the angle Θ in the counterclockwise direction from the X axis with respect to the predetermined X axis and the Y axis orthogonal thereto are calculated.

Then, the control unit 32 grasps the current position of the first rotating unit 14 and the second rotating unit 16, and then the second camera up to the image of the intruder 50 based on the distance R and the angle Θ. The rotation angle of the 1st rotation part 14 and the rotation angle of the 2nd rotation part 16 for moving (18) are calculated by a predetermined program. At this time, since the distortion characteristics of the fisheye lens 12 (or the convex mirror 90, the annular convex mirror 70, and the concave mirror 72) have a fixed value, the image of the first camera 10 is digitally displayed. It can be reconstructed like a normal picture by correcting with signal processing (DSP).

The rotation angle of the first rotation unit 14 and the rotation angle of the second rotation unit 16 calculated as described above are passed from the control unit 32 via the first drive signal output unit 36a and the second drive signal output unit 36b. It is transmitted to each servomotor (not shown) and driven.

FIG. 8 is an example of an image captured by the second camera 18 of the present invention moved according to the coordinates calculated in FIG. 7. That is, as shown in Figure 8, the image of the second camera 18 is caught by the intruder 50 in the center of the screen without distortion of the image.

Hereinafter, as described above, after the image of the intruder 50 is captured by the second camera 18, the process of changing the position of the second camera 18 according to the movement of the intruder 50 will be described in detail. Do it.

9 is an example of an image for determining a moving direction of the intruder 50 by comparing and calculating a plurality of images photographed by the second camera. First, when the intruder 50 moves in one direction, the change for each frame photographed is minute, and as shown in FIG. 9, the minute pixel data value changes between the nth frame and the n + 1th frame.

The controller 32 calculates the change of the pixel data value by comparing and calculating the n th frame and the n + 1 th frame. Next, the tracking vector 64 is calculated as shown in FIG. 10 by determining the change of the pixel data value (or the moving direction 62). That is, FIG. 10 is an example of an image for calculating the tracking vector 64 from the moving direction of FIG. 9. The tracking vector 64 includes the size (the length of the vector) and the angle data (the slope of the vector) of the movement direction 62 for the time corresponding to the interval of the frame.

Such size and angle data are converted into rotation angle data in which the first rotation unit 14 and the second rotation unit 16 should be further rotated in the control unit 32. The converted data is transmitted to each of the rotating units 14 and 16 through the driving signal output unit 36. Therefore, the second camera 18 moves slightly in the moving direction of the intruder 50.

Indeed, even if the intruder 50 walks a few steps, the angle at which the second camera 18 should be further rotated to track it is small.

Intruder tracking apparatus using the dual camera according to the present invention and the method, the video of each camera (10, 18) can be configured to always be recorded on the VCR or DVR (40) device, or the image of the first camera 10 Only the VCR or DVR 40 is always recorded in the device, and the image of the second camera 18 may be configured to be recorded only when the intruder 50 is present. In some cases, the video of the first and second cameras 10 and 18 may not be recorded, but may be recorded only when an intrusion is detected.

In addition, the VCR or DVR 40 or the image of the tracking driver 30 may be configured to transmit through the Internet. In this case, the central server apparatus (not shown) has the advantage of being able to simultaneously control a plurality of dual camera apparatuses 1.

In addition, the image storage means 40 of the present invention may be replaced by a hard disk, a CD-RW, a DVD-RAM, a magnetic recording tape, a memory stick, or the like in addition to a VCR and a DVR. In addition, although the dual camera device 1 according to the present invention is attached to the ceiling 20, it may be installed and used on an inclined surface, a wall surface, a bottom surface, etc. as necessary.

As described above, according to the intruder tracking device using the dual camera and the method according to the present invention, there is a feature that can always record, record and monitor the entire indoor or surveillance target area (security guard area).

In addition, when an intruder occurs in the guarded area, the camera automatically detects the intruder's moving position, so that the camera can capture the intruder's video while tracking the intruder, thereby enhancing security and the ability to cope with the situation. do.

Although the present invention has been described in connection with the above-mentioned preferred embodiments, various other modifications and variations may be made without departing from the spirit and scope of the invention. It is therefore contemplated that the appended claims will cover such modifications and variations as fall within the true scope of the invention.

Claims (10)

A first camera 10 whose one end is fixed and photographed in the axial direction; Optical means installed in the first camera 10 to map an image of a subject around the first camera 10 into a single image; A first rotating part (14) rotatably installed on an axis of the first camera (10); A second camera (18) mounted on the first rotating portion (14) and mounted on a second rotating portion (16) rotatable by a predetermined angle independently of the rotational movement of the first rotating portion (14); Determination means for determining the presence or absence of the intruder (50) from the plurality of image signals of the first camera (10) through the optical means; Calculating means for calculating coordinate data of the intruder 50 based on the video signal of the first camera 10 when a command indicating that the intruder 50 has invaded by the determining means is output; In order for the second camera 18 to photograph the intruder 50, a driving signal of the first rotating part 14 and the second rotating part 16 is generated based on the coordinate data, so that the corresponding rotating parts 14 and 16 are generated. A driving signal output unit 36 for transmitting to each; And Intruder tracking device using a dual camera, characterized in that it comprises an image storage means (40) for storing the images of the first camera (10) and the second camera (18). The method of claim 1, Computing means for comparing and calculating the plurality of video signals of the second camera (18); And Tracking vector calculating means for calculating a tracking vector representing a moving direction of the intruder 50 from the calculating means; More, The driving signal output unit 36 generates driving signals of the first rotating unit 14 and the second rotating unit 16 based on the tracking vector and transmits the generated driving signals to the corresponding rotating units 14 and 16, respectively. Intruder tracking device using dual camera. The intruder tracking device according to claim 1 or 2, wherein the video storage means (40) is a VCR or DVR device. The intruder tracking device using a dual camera according to claim 1 or 2, wherein the first camera (10) is fixed perpendicular to the ceiling (20) and is installed to photograph vertically downwards. The method of claim 1 or claim 2, wherein the first rotating part 14 is rotatable within a range of 0 ° ~ 360 °, the second rotating part 16 is characterized in that rotatable within a range of 0 ° ~ 90 ° Intruder tracking device using a dual camera. 2. An intruder tracking device according to claim 1, wherein said optical means is a fisheye lens (12). The method of claim 1, wherein the optical means, A column 94 having one end fixed thereto and the other end extending along an axis of the first rotating part 14 to protrude out of the first rotating part 14; A convex mirror 90 having one surface fixed to the column 94 and the other surface forming a mirror; A first camera 10 facing the convex mirror 90 and spaced apart from the convex mirror 90 by a predetermined distance; And One side is fixed to the column 94, the convex mirror 90 is installed therein, a double camera characterized in that consisting of a transparent case 92, the first camera 10 is fixed to one end Intruder tracking device used. The method of claim 1, wherein the optical means, An annular convex mirror (70) attached to the first camera (10) to reflect an image of a subject around the first camera (10); A concave mirror 72 installed to be spaced apart from the annular convex mirror 70 by a predetermined interval to reflect the image reflected by the annular convex mirror 70 to the first camera 10; And Intruder tracking device using a dual camera, characterized in that consisting of; a transparent case for supporting the annular convex mirror (70) and the concave mirror (72). Determining the presence of the intruder 50 from the plurality of image signals of the first camera 10 through the assembly of the fisheye lens 12, the convex mirror 90 or the annular convex mirror 70 and the concave mirror 72 ; Calculating coordinate data of the intruder 50 based on the image signal of the first camera 10 when a command indicating that the intruder 50 has invaded is output; In order for the second camera 18 on the second rotating part 16 mounted on the first rotating part 14 to photograph the intruder 50, the first rotating part 14 and the second rotating part based on the coordinate data. Generating driving signals of (16) and transmitting them to the corresponding rotating units (14, 16), respectively; Calculating a tracking vector representing a moving direction of the intruder by comparing a plurality of image signals of the second camera 18; And Generating driving signals of the first rotating part (14) and the second rotating part (16) based on the tracking vector and transmitting the driving signals to the corresponding rotating parts (14, 16), respectively; Intruder tracking method using a dual camera comprising a. 10. The method according to claim 9, further comprising the step of storing the video signal of the first camera (10) and the video signal of the second camera (16), respectively.