KR101049050B1 - Monitoring Camera System and Mothod for Controlling the Same - Google Patents

Abstract

The present invention relates to a security camera system and a control method thereof. The present invention includes a camera unit for photographing at least one reference member and a subject; Security camera system including a control unit for calculating the actual size of the subject using the size (image size of the reference member) and the image (subject image size) in the image of the reference member photographed by the camera unit; to provide.

Security, camera, reference member

Description

Monitoring Camera System and Mothod for Controlling the Same}

The present invention relates to a security camera system and a control method thereof, but not particularly limited to a city security camera system suitable for installation and use in an alley or a back road in the city center and a control method thereof.

In order to prevent accidents and accidents and to read and store recorded images at the time of accidents, CCTV cameras are increasingly installed in predetermined locations such as urban alleys and back roads in order to solve accidents.

In general, a security camera system is installed in the area to be photographed, the camera unit for photographing the area, the control unit connected to the camera unit and is normally installed at the site where the camera is installed, the number of the camera is installed The remote control unit is installed at a remote location of a kilo or tens of kilometers and connected to the control unit.

The control unit refers to a computer-based control device, is used to store the image taken by the camera unit and to change the shooting conditions, etc. in the camera unit, generally an industrial computer can be used. The remote control unit usually means a computer installed in a monitoring center located at a remote place such as a police department or a local government. In the case of the police station, the remote control unit monitors the site where the camera unit is installed in real time or reproduces the stored image in the event of an accident and grasps the site situation afterwards.

In order to prevent and solve an accident by using a security camera system, it is necessary to accurately identify at least the person and the vehicle that are taken from the image taken by the camera unit. That is, it is necessary to identify the impression wear including the face of the person photographed and the license plate of the vehicle.

Furthermore, in addition to the usual information such as impression wear including the face of the person photographed in the security camera system, if the key and the like can be known, it can be used more usefully when a crime occurs. However, the security camera system that is commonly used today can not calculate the height of the person photographed.

Patent Application No. 10-1992-011249 (Camcorder distance) as a method for calculating the height, height, width, etc. (hereinafter, collectively referred to as "size") of a photographed subject (person, vehicle, etc.) in the captured image. And height measuring method) and the like have been proposed. However, in this method, the distance and height of the subject are calculated by calculating the focal length and the imaging angle of the camera when the subject is photographed. Therefore, it is difficult to use when the focal length and the imaging angle of the camera is changed. In addition, there is a disadvantage that the calculation method also changes if the camera type is different. Therefore, in general, the security camera system is difficult to use this method because the focal length and the imaging angle of the camera changes depending on the situation.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned problems of the prior art, and an object of the present invention is to provide a security camera system and a method of controlling the same, which can obtain additional information such as a key from a photographed subject.

In order to achieve the above object, the present invention includes a camera unit for photographing at least one reference member and the subject; Security camera system including a control unit for calculating the actual size of the subject using the size (image size of the reference member) and the image (subject image size) in the image of the reference member photographed by the camera unit; to provide.

Here, the reference member includes a first reference member and a second reference member, and the control unit calculates an image size (weather reference member image size) of the virtual reference member using the first reference member and the second reference member. The actual size of the photographed subject may be calculated using the virtual reference member image size. In this case, the control unit may calculate the actual size of the subject a plurality of times, and calculate an average value using the calculated number of subject calculations.

According to another embodiment of the present invention, the present invention further comprises a movement means for moving the camera unit up, down, left, and right, the control unit is the camera using the exercise means if the reference member is not photographed for a predetermined time It provides a security camera system, characterized in that for moving the unit to a predetermined reference position.

According to still another embodiment of the present invention, there is provided a security camera system, characterized in that it further comprises a marker that is located on the extension line of the reference member and actually displayed on the road. Preferably, the control unit calculates the actual size of the subject by using the marker.

According to still another embodiment of the present invention, there is provided an image recording method comprising: a photographing step of photographing at least one reference member and a subject; An analysis step of analyzing the size (reference member image size) in the image of the reference member and the size (subject image size) in the image of the subject from the image photographed in the photographing step; And a calculating step of calculating the actual size of the subject by using the actual size of the reference member, the image size of the reference member, and the image size of the subject.

Here, photographing the first reference member and the second reference member in the photographing step; In the analyzing step, analyzing image sizes of the first reference member and the second reference member; In the calculating step, the image size of the virtual reference member (the image size of the weather reference member) is calculated using the image sizes of the first reference member and the second reference member, and the subject photographed using the image size of the virtual reference member. It is desirable to calculate the actual size of. In this case, it is preferable to calculate the actual size of the subject a plurality of times in the calculating step, and calculate an average value using the calculated number of subject calculations.

According to another embodiment of the present invention, the present invention further comprises a security camera step of moving the camera unit to a predetermined reference position if the reference member is not photographed for a predetermined time in the photographing step. Provides a control method.

According to still another embodiment of the present invention, the present invention provides a control method of a security camera system, characterized in that for photographing the marker that is located on the extension line of the reference member actually displayed on the road in the photographing step. It is preferable to calculate the actual size of the subject using the marker.

Security camera system according to the present invention and the control method thereof are as follows.

First, according to the present invention, since the height of the photographed person can be calculated accurately, there is an advantage in that information about the suspect can be obtained more accurately when a crime occurs.

Secondly, according to the present invention, there is an advantage that the height of the photographed person can be accurately calculated regardless of the type of camera and the state of the lens.

Third, according to the present invention, there is an advantage that the camera can be automatically restored to the original position when the camera is changed from the originally installed position.

Hereinafter, a preferred embodiment of a security camera system and a control method according to the present invention.

First, referring to Figure 1, the overall configuration of a preferred embodiment of a security camera system according to the present invention will be described.

For convenience, this embodiment will be described by taking an example of the configuration of the CCTV camera unit, the camera unit itself is a well-known technology, so it will be briefly described as limited to the parts necessary for the description of the present invention.

The camera unit 100 is connected to the control unit 200. Of course, the control unit 100 may be connected to the remote control unit 400 through a communication network 300 such as the Internet.

The camera unit 100 is preferably a megapixel camera of 1 million pixels or more capable of capturing a high resolution image. The camera unit 100 basically includes a lens, a shutter, an image sensor (image pickup device such as CCD or CMOS), and the like, and the lens, shutter, image sensor, etc. are appropriately controlled by a processor. Of course, the camera unit 100 has an input and output interface that is connected to the outside, that is, the control unit 200. The camera unit is a known technology and is not related to the subject matter of the present invention, so a detailed description thereof will be omitted.

The remote control unit 400 generally means a computer installed in a monitoring center located at a remote place such as a police department, a local government, or a private security company. The National Police Agency monitors the scene in which the camera unit 100 is installed in real time using an image photographed by the camera unit 100 in the remote control unit 400 or post-mortem situation by using the stored image when an accident occurs. Figure out.

The control unit 200 will be described as follows.

The control unit 200 may be a computer-based control device, used to store the image taken by the camera unit 100 and to change the shooting conditions, etc. in the camera unit 100, generally an industrial computer can be used have. Accordingly, the control unit 200 may include a central processing unit (processor), a memory unit, an auxiliary memory device, an input unit, an output unit, and the like. The memory unit may include various programs for operating the camera, and an SDRAM or a fresh memory may be used. In addition, the auxiliary memory device may be used a variety of memory cards, hard disks, USB, etc., and stores the image taken by the camera unit 100 and the like. The input unit performs various operations by the user, and generally employs a graphical user interface. The output unit displays various contents to the user, and it is common to use TPI LCD. In addition, the central processing unit preferably has a camera interface connected to the camera unit 100, and a remote interface connected to the remote control unit 400. The input / output interface of the camera unit 100 may be any of various methods such as Ethernet, USB, IEEE1394, Camera Link, LVDS, etc. Therefore, the camera interface of the control unit 200 may be selectively or collectively connected with these interfaces. Interface that can be connected. Therefore, the present invention is not limited to the specific interface of the camera unit 100 can be connected to various types of cameras.

On the other hand, the camera unit 100 and the control unit 200 is generally installed adjacent to the place where the normal road 2 to be photographed. In addition, a support member (pole) 44 is installed at one side of the road 2 in a vertical direction, and a connection member (female) 46 is connected to the support member substantially horizontally to the other end of the connection member 46. In general, the camera unit 100 is installed.

Meanwhile, in the present invention, at least one reference member 62 having a predetermined size (length) is installed at the place where the camera unit 100 is photographed. The size of the reference member 62 can be arbitrarily determined. And, the reference member 62 is located within the shooting range of the camera unit 100, it is usually preferably installed next to the road (2). The reference member 62 may use a separate member, and may draw a reference member having a predetermined size, such as a power pole or a wall that is usually installed next to a road.

Referring to FIGS. 2 and 7, a method of calculating an actual size of a subject in an embodiment of the present invention will be described below. FIG. 2 shows one image actually captured, that is, a praim (F). The portion G1 indicated by a dotted line is shown for convenience of description and means a reference line extending horizontally from the installation position of the reference member 62.

First, the camera unit 100 photographs a predetermined area (S3). Of course, the camera unit 100 should photograph at least a subject (eg, a person) 82 and a reference member 62 whose size is to be known. . The camera unit continuously photographs a plurality of frames, but calculates a key of the subject 82 by using an image (frame) F that captures a moment when the subject 82 passes the reference line G1.

On the other hand, in order to accurately find the factors for calculating the subject size, it is preferable to properly correct the photographed image F through filtering or the like. (S5) By analyzing the last photographed image (S7), the actual size of the subject (S9)

Calculating the actual size of the subject 82 will now be described in detail.

For convenience, the actual size of the reference member 62 is Hr, the size in the image of the reference member 62 photographed by the camera unit 100 (the reference member image size) is H1, the size in the image of the photographed subject 82. (Subject image size) is called h, and the size of the subject 82 ("subject calculation size") calculated using these factors is called hr.

Then, the relationship of H1: Hr = h: hr is established, so that the calculated size hr of the subject becomes (Hr * h) / H1. However, since the actual size Hr of the reference member 62 is known in advance, the image size H1 and the subject image size h of the reference member 62 may be obtained by analyzing the captured image. The size of the subject 82 may be calculated at. The reference member image size H1 and the subject image size h can be obtained in various ways, and since it is simple to calculate the number of pixels (pixels), it is preferable to use them. When calculating the size of an image using the number of pixels, pixels in the image are generally determined horizontally and vertically. When the image to be calculated is not in the form of a horizontal or vertical line, a trigonometric function is used to determine the number of horizontal pixels and the number of vertical pixels. It is preferable to calculate suitably using.

There are various methods of image processing techniques for identifying the reference member 62 and the subject 82 in the captured image, and any method that can identify the reference member 62 and the subject 82 may be used. It is possible. For example, the reference member 62 and the subject 82 may be found in an image to obtain the size of the image by analyzing the original image using a difference of Gaussian (Dog) and a Laplacian of Gaussian (LoG) processing method. .

To accurately calculate the actual size of the subject 82 using the above-described embodiment, as shown in FIG. 4A, the camera unit 100 is installed perpendicular to the direction in which the subject 82 travels. It is preferable. This is because the reference line G1 is positioned perpendicular to the photographing direction of the camera unit 100 so as to be installed in this way. However, it is often difficult to actually install the camera unit 100 in this way. This is because, in order to install this, the length of the connecting member 46 connected to the strut member 44 should be long, because the length of the connecting member 46 is not aesthetically good and the shaking of the camera unit 100 also increases. .

Therefore, as shown in FIG. 6B, when the actual camera unit 100 is installed, the camera unit 100 is normally installed to be inclined at a predetermined angle and a moving direction of the subject 82. However, in this case, since the actual reference line G1 and the reference line Gc in the captured image are different, an error may occur in the calculated actual size. Therefore, it is desirable to appropriately correct the calculated size he of the calculated subject. However, as a result of the experiments of the present inventors, instead of properly correcting the calculated size he of the subject, the marking portion 22 is provided at a predetermined position on the road surface of the road-the portion corresponding to the reference line G, and the marking Even if the actual size of the subject was calculated on the basis of the unit 22, there was no big error.

Referring to FIG. 5, the calculation of the actual subject size is as follows.

As a result of the experiment, the height of a person having a height of 172 cm was calculated as 173.9 cm using the above-described embodiment. That is, according to the present invention, the height of a person can be calculated relatively accurately with a very small error.

With reference to FIG. 6, another embodiment of a security camera system and a control method thereof according to the present invention will be described.

This embodiment also has the same basic principle as the above-described embodiment. However, in the above-described embodiment, an image (frame) F (see FIGS. 2 and 5) when the subject 82 passes the reference line G1 should be used. In the security camera system usually shoots 5 frames per second, but the main subject to be recognized in the security camera system is moving like a human, there may be no frame that the toe of the person exactly passed the reference line (G1). In view of such a situation, it is conceivable to install a plurality of reference members, but this method is complicated. The present embodiment also provides a method for accurately calculating the size of the subject 82 even in this case.

In this embodiment, a second reference member 64 is additionally provided in addition to one reference member (called a "first reference member") 62. Then, the reference member is not actually installed between the first reference member 62 and the second reference member 64, and if installed, the first reference member 62 determines how much the actual size and the size in the image are. ) And the second reference member 64 using the actual size and the image size.

That is, when the subject 82 passes one of the positions G (hereinafter, referred to as a "virtual reference line") between the first reference member 62 and the second reference member 64, the captured image ( Frame) —these images must be present—to calculate the actual size of the subject 82.

Referring to Figures 7 and 8, it will be described in detail as follows.

If it is assumed that the virtual reference member 66 is actually installed in the virtual reference line G, the calculated size hr of the subject 82 becomes (Her * h) / He. Here, Her is the actual size of the virtual reference member 62, He is the image size of the virtual reference member 62 photographed by the camera unit 100, h is the image size of the photographed subject 82.

Here, the actual size Her of the virtual reference member 66 and the size He in the image are calculated using the first reference member 62 and the second reference member 62. Referring to Figure 8, it is described as follows.

A first reference member 62 and a second reference member 62 having the same actual size are provided. Even though the actual size of the first reference member 62 and the second reference member 62 is the same, the closer to the camera unit 100, the larger the image size will be. The size He in the image of the virtual reference member 66 having the same actual size as the actual size of the first reference member 62 and the second reference member 62 may be obtained as follows.

 Since the relationship of (He-H2) / Le = (H1-H2) / L is established, the magnitude He in the image of the virtual reference member 66 can be obtained.

9 actually calculates the height of the person 82 using this embodiment. As can be seen in Figure 9, the actual human height is 175cm, the calculated height is 175.24cm, it can be seen that the error is only 0.24cm. As a result of many experiments, the error tends to increase as the virtual reference line G is farther from the first reference line G1 and the second reference line G2, but the error range is less than about 1%.

On the other hand, when two reference members are used, a plurality of images (frames) are used to calculate the actual size of the subject in each frame a plurality of times. It is also possible to set the average value of the calculated actual size of the subject several times as the actual size of the subject.

Meanwhile, a method for reducing the error between the calculated size and the actual size is as follows. It is desirable to select an image when one foot is off the ground or the center of gravity is moved to one foot. In addition, in principle, the size of the reference member can be arbitrarily determined, but the approximate size of the subject to measure the height of the reference member, for example, when measuring mainly the height of the person, can reduce the error to about 150-180 cm. In addition, since a typical road is designed to have a high central portion and a low left and right side, correction may be made in consideration of the inclination of the road if necessary.

Another embodiment of a security camera system and a control method thereof according to the present invention will be described.

The present embodiment relates to another embodiment using the reference members 62 and 64 included in the captured image. In the security camera system, the position of the camera unit 100 may be changed. That is, the direction of the lens of the camera unit 100 may be changed manually or automatically to change the area photographed by the camera unit 100. In this case, the camera unit 100 preferably returns to the originally set position, but in many cases, the camera unit 100 cannot return to the original position by mistake.

Therefore, in the present embodiment, if the reference members 62 and 64 are not photographed for a predetermined time in the control unit 200, the camera unit 100 is moved to a predetermined reference position. Of course, for this purpose, it further includes a movement means (not shown) such as a pan tilt (Pan Tilt) that can move the camera unit 100 up, down, left and right.

1 is a block diagram showing the overall configuration of an embodiment of a security camera system according to the present invention

FIG. 2 is a diagram illustrating an image (frame) captured by the security camera system of FIG.

3 is a flowchart illustrating an embodiment of a control method of a security camera system according to the present invention.

Figure 4 is a block diagram showing the overall configuration of another embodiment of a security camera system according to the present invention

5 is a diagram illustrating an image (frame) captured by the security camera system of FIG.

Figure 6 is a block diagram showing another embodiment of a security camera system according to the present invention

7 and 8 are graphs for explaining the principle of the embodiment of FIG.

9 is a diagram illustrating an image (frame) captured by the security camera system of FIG.

Claims (4)

A camera unit for photographing a subject; At least one reference member separately installed within a shooting range of the camera unit; A control unit for calculating the actual size of the subject by using the actual size of the reference member, the size in the image of the reference member photographed by the camera unit (the reference member image size), and the size in the image of the subject (the subject image size); Security camera system including. The apparatus of claim 1, wherein the reference member includes a first reference member and a second reference member, and the control unit uses an image size of a virtual reference member using the first reference member and the second reference member. Size), and calculates the actual size of the photographed subject by using the virtual reference member image size. A photographing step of photographing at least one reference member and a subject separately installed within the photographing range of the camera unit; An analysis step of analyzing the size (reference member image size) in the image of the reference member and the size (subject image size) in the image of the subject from the image photographed in the photographing step; And calculating the actual size of the subject by using the actual size of the reference member, the image size of the reference member, and the image size of the subject. 4. The method of claim 3, further comprising: photographing the first reference member and the second reference member in the photographing step; In the analyzing step, analyzing image sizes of the first reference member and the second reference member; In the calculating step, the image size of the virtual reference member (the image size of the weather reference member) is calculated using the image sizes of the first reference member and the second reference member, and the subject photographed using the image size of the virtual reference member. The method of controlling a security camera system, characterized in that for calculating the actual size of the.

Images