KR100359342B1 - System for automatic installation of detection area using panning, tilting and zooming transformation of cctv cameras - Google Patents

Abstract

The present invention relates to a detection area automatic setting system according to Panning, Tilting, Zooming conversion of CCTV cameras. The present invention provides a panning, tilting, and zooming of a CCTV camera in order to add an image detector function to automatically collect traffic information to a CCTV camera that is installed on a highway, a city intersection, a national road, or a road to be constructed in the future. It is a detection area automatic setting system that collects traffic information such as traffic volume, average speed, and occupancy in the detection area for each lane by automatically setting the detection area for each lane of the road from the video from the CCTV camera whenever the conversion is performed.

Description

SYSTEM FOR AUTOMATIC INSTALLATION OF DETECTION AREA USING PANNING, TILTING AND ZOOMING TRANSFORMATION OF CCTV CAMERAS}

The present invention relates to a detection area automatic setting system according to Panning, Tilting, Zooming conversion of CCTV cameras. More specifically, in order to add an image detector function to automatically collect traffic information on a CCTV camera installed on a highway, an urban intersection, a national road, or a new road, the Panning, Tilting, When the zooming conversion is performed, the CCTV camera that automatically collects traffic information such as traffic volume, vehicle average speed, and occupancy in the detection area for each lane by automatically setting the detection area for each lane on the road from the video from the CCTV camera. The present invention relates to a detection area automatic setting system according to panning, tilting, and zooming conversion.

In general, image detectors are installed and operated as a method of automatically collecting traffic information of a road point using only existing image information, but this should be installed by fixing the direction of the CCD camera used for the image detector. In addition, the range of the image coming into the camera also reflects a very narrow point of about 20 meters to 30 meters of the road ahead, there is a problem that it is impossible to collect traffic information of the point outside the range.

In addition, since the direction of attention of the CCD camera of the image detector is fixed, traffic information can be collected only at the corresponding point when the detection zone is set once when the image detector is first installed, and it is impossible to grasp the traffic situation of the front or rear point. The disadvantage is that.

On the other hand, in the case of CCTV cameras installed on the road, unlike conventional video detectors, a wide range of images from 2km to 3km are received, and the camera is directed to a desired point using panning, tilting, and zooming functions of CCTV cameras. While there is an advantage in that it can grasp the traffic situation of the branch in real time, it does not collect traffic information automatically like a video detector, but only the images from CCTV cameras. The situation is grasping.

As described above, the existing image detector has the advantage of automatically collecting traffic information, while the point where the traffic information can be grasped is very narrow. There is a growing problem, and CCTV cameras are easy to observe a wide range of points, but because the traffic information center personnel are visually recognized, the accuracy of the traffic information is greatly reduced, so it is not possible to respond quickly to congestion or accidents. Have

Therefore, the present invention has been made to solve the above problems, the purpose is to automatically collect traffic information in the CCTV cameras that are installed in highways, downtown intersections, national roads, etc. or will be installed in the future due to the construction of new roads. By adding the image detector function, it is possible to automatically grasp the traffic information that was currently only understood by the naked eye of the personnel of the traffic information center, which can greatly improve the accuracy of the traffic information, and promptly deal with it in case of traffic jams or accidents. As a result, a reliable system can be provided.

In addition, another object of the present invention is to provide an economical road operation because there is no need to install an image detector at the point where the existing CCTV camera is installed because one CCTV camera can observe a wide range of points.

1 is a block diagram schematically showing the configuration of a detection area automatic setting system by Panning, Tilting, Zooming conversion of a CCTV camera according to the present invention

Figure 2 is a block diagram showing the configuration of the element value calculation unit of the automatic detection area setting system according to the Panning, Tilting, Zooming transformation of the CCTV camera according to the present invention

Figure 3 is a block diagram showing the configuration of the communication unit of the detection area automatic setting system according to the Panning, Tilting, Zooming conversion of the CCTV camera according to the present invention

4 is an exemplary view of a straight line for each lane for obtaining an equation of a straight line connecting each lane according to the present invention.

5 is a photograph showing a CCTV camera image (Gyeongheung Expressway Giheung Interchange up line image, Panning value: 45, Tilting value: 130, Zooming value: 246) initially set the detection area according to the present invention

6 is a flowchart illustrating a procedure for automatically setting a detection area when a CCTV camera is panning according to the present invention.

7 is a photograph showing the resulting image by the procedure of FIG.

8 is a flowchart illustrating a procedure of automatically setting a detection area when a CCTV camera is tilted according to the present invention.

FIG. 9 is a photograph showing a resultant image (the Gyeonggi Expressway Giheung Interchange up line image, Panning value: 65, Tilting value: 120, Zooming value: 246) as a result of the procedure of FIG.

<Explanation of symbols for main parts of the drawings>

110: CCTV camera 120: video input unit

130: element value calculation unit 140: communication unit

150: coordinate transformation calculation unit 160: correction unit

170: preset setting unit

Technical configuration of the present invention for achieving the above object,

CCTV cameras installed on the road to shoot road video,

Image input unit for compressing the image input from the direction of attention of the CCTV camera to transmit to the traffic information center,

An element value calculator for calculating three element values of panning, tilting, and zooming values of the CCTV camera when the CCTV camera is panning, tilting, and zooming;

A communication unit for transmitting each element value calculated from the element value calculation unit to a traffic information center;

A coordinate transformation calculation unit installed in a traffic information center and calculating coordinate transformation in three-dimensional space by a coordinate transformation algorithm in three-dimensional space using each element value transmitted from the communication unit as a parameter;

A correction unit which is installed in a traffic information center, and corrects a predetermined detection method so that the detection area is automatically set in consideration of the mechanical error of the CCTV camera and then accurately located in each lane of the road;

It is installed in the traffic information center, and sets the main observation point and the occasional observation point between the front and rear viewing direction maximum points that the CCTV camera can observe, and in advance, the value of each element of the CCTV camera watching these points in advance. It characterized in that it comprises a preset setting unit for setting the coordinates of the detection area according to the setting.

Herein, the element value calculating unit

An A / D converter which reads a variable resistance value inside the CCTV camera and converts it into a digital signal;

A CPU for calculating a current value for panning, tilting and zooming input from the A / D converter, and outputting a control signal to a corresponding driver when control of panning, tilting and zooming of the CCTV camera is input from the outside;

A Pan / Tilt driver for outputting a driver signal for panning and tilting a CCTV camera according to a control signal input from the CPU;

A zoom driver for outputting a driver signal for zooming a CCTV camera according to a control signal input from the CPU;

And a communication driver for transmitting each element value calculated by the communication unit or receiving a control signal input from the traffic information center.

Here, the predetermined correction method of the correction unit

The image is divided into three parts, the upper part, the lower part, and the center part, centered on the set detection area, and the image contrast value is analyzed in the horizontal direction.

Here, the preset setting unit

By using the preset function, the gaze direction of the CCTV camera is automatically moved to the corresponding point, and the detection area is automatically set by reading the coordinates of the preset detection area for each preset.

The preset setting unit

If the traffic information is collected at a point other than a preset point, a coordinate transformation is calculated based on a preset closest to the panning, tilting, and zooming values of the CCTV camera.

The communication unit

Pass each element value through RS-232C port,

The video and data elements of the CCTV camera are multiplexed and transmitted through the optical communication system.

Hereinafter, the configuration of the automatic detection area automatic setting system according to Panning, Tilting, and Zooming conversion of a CCTV camera according to the present invention will be described in detail with reference to FIGS. 1 to 3.

1 is a block diagram schematically showing the configuration of the automatic detection area setting system by Panning, Tilting, Zooming conversion of the CCTV camera according to the present invention, Figure 2 is a conversion of Panning, Tilting, Zooming of the CCTV camera according to the present invention 3 is a block diagram showing the configuration of the element value calculator of the automatic detection area according to the present invention, Figure 3 is a block diagram showing the configuration of the communication unit of the automatic detection system according to the Panning, Tilting, Zooming conversion of the CCTV camera according to the present invention .

1 to 3, the detection area automatic setting system 100 according to the Panning, Tilting, Zooming conversion of the CCTV camera according to the present invention, the CCTV camera 110, the image input unit 120, the element value The calculator 130, the communication unit 140, the coordinate transformation calculator 150, the correction unit 160, and the preset setting unit 170 are configured.

First, the CCTV camera 110 is installed on the road to take a road image, the image input unit 120 compresses the image input in the direction of attention of the CCTV camera 110 by a predetermined compression method to transmit to the traffic information center.

And the element value calculator 130 is shown in Figure 2 to calculate the three element values of the panning value, tilting value, zooming value of the CCTV camera 110 when the CCTV camera 110 is panning, tilting, zooming As shown, the A / D converter 131, the CPU 132, the Pan / Tilt driver 133, the Zoom driver 134, the communication driver 135 and the power driver 136.

The A / D converter 131 reads the variable resistance value inside the CCTV camera 110 and converts the variable resistance into a digital signal, and the CPU 132 presents the current for panning, tilting, and zooming input from the A / D converter 131. It calculates a value and outputs a control signal to a corresponding driver when control for panning, tilting and zooming of the CCTV camera 110 is input from the traffic information center.

The Pan / Tilt driver 133 outputs a driver signal for panning and tilting the CCTV camera 110 according to the control signal input from the CPU 132, and the zoom driver 134 outputs a control signal input from the CPU 132. According to the driver outputs a zoom signal for zooming the CCTV camera 110, the communication driver 135 transmits each element value calculated by the communication unit 140, or receives a control signal input from the traffic information center, power driver 136 supplies power to the CCTV camera 110.

In addition, the communication unit 140 transmits each element value calculated from the element value calculating unit 130 to the traffic information center. Herein, the communication unit 140 transmits each element value through the RS-232C port as shown in FIG. 3, and displays the video link and the data link of the image and each element value of the CCTV camera 110 through the optical communication system. Deliver by multiplexing

In addition, the coordinate transformation calculation unit 150 is installed in the traffic information center, and calculates the coordinate transformation in the three-dimensional space by the coordinate transformation algorithm in the three-dimensional space using each element value transmitted from the communication unit 140 as a parameter. .

In addition, the correction unit 160 is installed in the traffic information center, and after the automatic detection area is set in consideration of the mechanical error of the CCTV camera 110, and corrects it in a predetermined correction method so that it can be accurately located in each lane of the road. Here, the predetermined correction method is divided into three parts of the upper part, the lower part, and the center part around the center of the set detection area, and analyzes the contrast value of the image in the horizontal direction and then corrects it based on this.

In addition, the preset setting unit 160 is installed in the traffic information center, by setting the main observation point and occasional observation point between the maximum point of the front and rear gaze direction that the CCTV camera 110 can observe Each element value of the CCTV camera 110 looking at these points is set in advance to store the coordinates of the detection area accordingly. In addition, the preset setting unit 170 automatically moves the gaze direction of the CCTV camera 110 to the corresponding point by using the preset function, and reads the coordinates of the preset detection area for each preset to automatically set the detection area. do. In addition, when the preset setting unit 170 wants to collect traffic information at a point other than a preset point, the coordinate conversion is performed based on a preset closest to the panning value, tilting value, and zooming value of the corresponding CCTV camera 110. Calculate

Hereinafter, the operation of the detection area automatic setting system according to Panning, Tilting, and Zooming conversion of a CCTV camera according to the present invention will be described in detail with reference to FIGS. 4 to 9.

4 is an exemplary diagram of a straight line for each lane for obtaining an equation of a straight line connecting each lane according to the present invention, and FIG. 5 is a CCTV camera image (the Gyeongbu Expressway Giheung Interchange Upward Line image) initially setting a detection area according to the present invention. , Panning value: 45, Tilting value: 130, Zooming value: 246), Figure 6 is a flow chart illustrating a procedure for automatically setting the detection area when the CCTV camera panning in accordance with the present invention, Figure 7 FIG. 8 is a photograph showing the resultant image by the procedure of FIG. 6, and FIG. 8 is a flowchart illustrating a procedure for automatically setting a detection area when the CCTV camera is tilted according to the present invention, and FIG. 9 is a resultant image by the procedure of FIG. This is a picture showing upstream image of Giheung Interchange of Gyeongbu Expressway, Panning value: 65, Tilting value: 120, Zooming value: 246.

First, when the CCTV camera 110 is panning, tilting, and zooming by a worker in the traffic information center, the element value calculating unit 130 converts the variable resistance value into the CCTV camera 110 in the A / D converter 131. Read through and calculate three element values of panning value, tilting value, and zooming value from 0 to 255, respectively, and then the coordinate conversion calculator 150 located in the traffic information center through the communication driver 135 and the communication unit 140. To send).

Then, the coordinate conversion calculation unit 150 automatically sets the detection area with the panning, tilting and zooming values of the CCTV camera 110 received from the communication unit 140. The basic direction of the algorithm is the transmitted CCTV camera 110. The coordinate transformation in three-dimensional space is calculated using the Panning, Tilting, and Zooming values of.

The coordinate value of one point projected on the CCTV camera 110 is called P ( x , y ), Z is called the parallel depth of XY plane, and θ, Ψ and Φ are respectively The angles of the X, Y, and Z axes, T _X , T _Y , and T _Z are called displacements of the X, Y, and Z axes, respectively, s is the zooming factor value, and f is the focal length. Coordinates converted in space may be expressed by Equation 1 below.

In Equation 1, θ, Ψ, and Φ are sufficiently small, _Z is much larger than T _X , T _Y , and T _Z , and Ψ and θ are smaller than x and y, respectively. Since these are very small compared to the f value, Equation 1 can be arranged as Equation 2 below.

x '= s (x + Ψ f)

y '= s (y- Φf)

In Equation 2, Ψ and θ are respectively related to the panning value and the tilting value of the CCTV camera 110, and f is multiplied by Ψ and θ to mean the correlation with the position of the CCTV camera 110.

However, because the CCTV camera 110 is located on the side of the road, the distance between each lane is not constant depending on the perspective. Therefore, in order to find the exact coordinate value of each lane during panning, the equation of the straight line connecting each lane should be obtained by using the vertical and horizontal angles of this straight line.

4 represents a straight line connecting each lane.

When the CCTV camera 110 is located at the i th center on the side of the road, and the angle between the j th lane and the j + 1 th lane has a deviation of η, the deviation ratio of each lane is expressed by Equation 3 below. I can express it.

Here j means the number of lanes, L _w means the actual width of the lane, C _h means the height of the CCTV camera (110).

Equation 3 is a proportional expression for calculating that the width becomes narrower as the distance from the view. By applying this proportional formula, the ratio of lane width is maintained at a constant ratio even when the angle is changed. Assuming that θ _x is the panning angle of the CCTV camera 110, the conversion angle to lane 0 can be expressed as Equation 4 below, and the conversion angle to the i th lane can be expressed as Equation 5 below. There is a number.

In Equation 5, i is i = 1, 2, 3 when there are four lanes. This formula is made in consideration of the ratio of the front lane and the rear lane so that the width of the lane narrows at a constant ratio as it moves away from the field of view.

When P (ξ, ζ) is called the coordinate value of the intersection point of the straight lines connecting each lane, and P (x ₁ , y ₁ ) is the coordinate value of one of the initially set detection areas, the CCTV camera 110 According to the panning value, when the coordinates of the detection area for each lane are P (x ', y'), P (x ', y') can be expressed by Equation 6.

That is, as shown in FIGS. 5 to 7 when a worker of the traffic information center inputs a panning value of the CCTV camera 110 to automatically set the detection area when the CCTV camera 110 pans, an input confirmation response After extracting the panning factor through the coordinate conversion calculation unit 150 calculates the transformed coordinates of the detection area according to Equation 6 above, and automatically set the detection area conversion position, the CCTV camera 110 is replaced with a new one. Move to the area.

In addition, when the CCTV camera 110 performs a tilting conversion, the depth of the image should be basically calculated. This depth can be calculated with the height of the CCTV camera 110 and the distance from the camera.

θ _x is called the tilting angle of the CCTV camera 110, δ is the actual distance from the CCTV camera 110 to the center point of the image, and when the CCTV camera 110 is tilted, d is the actual moving distance. The equation to be calculated can be calculated by the equation (7).

That is, as shown in FIGS. 6, 8, and 9, when a worker of the traffic information center inputs the tilting value of the CCTV camera 110 to automatically set the detection area when the CCTV camera 110 tilts. After extracting the Tilting factor through the input confirmation response, the coordinate transformation calculation unit 150 calculates the transformation coordinates of the detection area according to Equation 7 above, automatically sets the detection area conversion position, and then sets the CCTV camera ( Move 110) to the new area.

When also the CCTV camera 110 have been called if the Zooming transformation has a Z Zoom magnification, Z value on the basis of the center of the image is to be a large magnified image than 1, Z value is a small, reduced image than the first. That is, the coordinate value of the detection area can be converted or reduced according to the zoom magnification. P (x _c , y _c ) is called the center coordinate of the image, the coordinate value before zooming transformation is P (x _i , y _i ), and the coordinate value after zooming transformation is P (x _k , y _k ) In this case, x _k and y _k can be obtained from Equation 8.

As described above, the coordinate conversion calculation unit 150 may set the detection area by calculating the coordinate value conversion of the detection area by the algorithm of the above-described method whenever there is Panning, Tilting, or Zooming conversion in the CCTV camera 110. .

When the setting of the detection area is completed through the coordinate conversion calculation unit 150, the preset setting unit 170 is generated by the Pan / Tilt driver 133 and the Zoom driver 134 of the element value calculation unit 130. In order to minimize the influence on the automatic detection area setting due to mechanical error, the main observation point and the occasional observation point are set between the maximum and front viewing direction points that the CCTV camera 110 can observe, The panning value, tilting value, and zooming value of the CCTV camera 110 to be observed are initially set in advance, and the coordinates of the detection area are stored in a file accordingly.

At this time, each preset point is stored by giving a focus number of the CCTV camera 110, and when the working personnel of the traffic information center selects the focus number of the main observation point and the occasional observation point, the CCTV camera 110 as the corresponding preset point. Panning, tilting, and zooming are performed, and the detection area is set in the image input unit 120 by reading coordinates of the detection area from the file.

If the worker of the traffic information center rotates the CCTV camera 110 to a point outside the preset point by the focus number of the CCTV camera 110, the panning value, tilting value, and zooming of the current CCTV camera 110 The value is compared with a preset point set in advance to search for the closest preset point and calculates and sets the coordinates of the detection area by using the algorithm of the coordinate conversion calculator 150 based on the preset point.

On the other hand, when the CCTV camera 110 is slightly out of the position initially set by the wind or the external environment, an error occurs in the image coordinate value of the image input unit 120. In this case, the detection unit 160 detects the area. Re-install in the car. Since the optimal position where the detection area is located is located between the lanes separating the lanes and lanes, when the position of the CCTV camera 110 is changed by the external environment, the detection area set by the preset setting unit 170 is Even if the coordinates are read from the file, an error occurs. In most cases, the detection area is set on the lane where the lanes are divided.

At this time, the correction unit 160 analyzes the contrast value of the image in the horizontal direction by dividing the three parts of the upper part, the lower part and the center part around the center of the set detection area. In general, the contrast of the road and the contrast of the lane marked on the road are significantly different from each other. For example, when the right part of the currently set detection area is located at the coordinate with the lane contrast value, the coordinates of the current detection area are detected in the negative direction of the X axis, that is, the left direction and the positive direction of the Y axis, that is, the lower direction. By moving the coordinates of, the detection area is corrected so that the detection area is located between the lanes separating lanes. Also, when the left part of the currently detected detection area is located at the coordinate with the lane contrast value, the coordinates of the detection area are moved in the positive direction of the X axis, that is, the right direction and the negative direction of the Y axis, that is, the upper direction. You can correct it by moving it.

Therefore, even if a worker in the traffic information center moves the direction of attention of the CCTV camera in the desired direction, the detection area at the corresponding point can be automatically set by using the preset function and coordinate transformation function in the 3D space. By providing the traffic information of the branch in real time, it is possible to determine the exact traffic situation.

As described above, according to the detection area automatic setting system according to the Panning, Tilting, and Zooming transformation of the CCTV camera according to the present invention, it is installed in highways, city intersections, national roads, etc. By adding a video detector function to automatically collect traffic information on CCTV cameras, it is possible to quickly and quickly identify traffic information that was only visible to the personnel of the current traffic information center, thereby greatly improving the accuracy of the traffic information. It is possible to respond quickly and reliably in case of congestion or accident.

In addition, since a single CCTV camera can observe a wide range of points, there is less need to install an image detector at the point where the existing CCTV camera is installed, so there is an advantage in that it is economical to operate the road.

Claims (6)

CCTV cameras installed on the road to shoot road video, Image input unit for compressing the image input from the direction of attention of the CCTV camera to transmit to the traffic information center, An element value calculator for calculating three element values of panning, tilting, and zooming values of the CCTV camera when the CCTV camera is panning, tilting, and zooming; A communication unit for transmitting each element value calculated from the element value calculation unit to a traffic information center; A coordinate transformation calculation unit installed in a traffic information center and calculating coordinate transformation in three-dimensional space by a coordinate transformation algorithm in three-dimensional space using each element value transmitted from the communication unit as a parameter; A correction unit which is installed in a traffic information center, and corrects a predetermined detection method so that the detection area is automatically set in consideration of the mechanical error of the CCTV camera and then accurately located in each lane of the road; It is installed in the traffic information center, and sets the main observation point and the occasional observation point between the front and rear viewing direction maximum points that the CCTV camera can observe, and in advance, the value of each element of the CCTV camera watching these points in advance. Automatic detection system according to the Panning, Tilting, Zooming conversion of the CCTV camera, characterized in that it comprises a preset setting unit for storing the coordinates of the detection area according to the setting. The method of claim 1, The element value calculating unit An A / D converter which reads a variable resistance value inside the CCTV camera and converts it into a digital signal; A CPU for calculating a current value for panning, tilting and zooming input from the A / D converter, and outputting a control signal to a corresponding driver when control of panning, tilting and zooming of the CCTV camera is input from the outside; A Pan / Tilt driver for outputting a driver signal for panning and tilting a CCTV camera according to a control signal input from the CPU; A zoom driver for outputting a driver signal for zooming a CCTV camera according to a control signal input from the CPU; And a communication driver for transmitting each element value calculated by the communication unit or receiving a control signal input from a traffic information center. The method of claim 1, The predetermined correction method of the correction unit Panning, tilting, and zooming conversion of CCTV cameras, characterized by analyzing the contrast value of the image in the horizontal direction by dividing it into three parts, the upper part, the lower part, and the center part, centered on the set detection area. Detection zone automatic setup system The method of claim 1, The preset setting unit Panning, Tilting of CCTV camera, characterized in that the gaze direction of the CCTV camera is automatically moved to the corresponding point by using the preset function, and the detection area is automatically set by reading the coordinates of the preset detection area for each preset. Detection zone automatic setup system according to zooming and zooming. The method of claim 4, wherein The preset setting unit Panning of CCTV cameras, characterized in that the coordinate transformation is calculated based on a preset closest to the panning value, tilting value, and zooming value of the CCTV camera when the traffic information is collected at a point other than a preset point. Detection zone automatic setting system according to conversion, tilting and zooming. The method of claim 1, The communication unit Pass each element value through RS-232C port, Auto detection system according to Panning, Tilting, Zooming conversion of the CCTV camera, characterized in that the video link and data link of the CCTV camera through the optical communication system and each element value is transmitted by multiplexing.