KR102060113B1 - System and method for performing calibration - Google Patents

Abstract

According to the present invention, a calibration system comprises: at least four patches placed on a roadway; a camera filming a road footage; and a calibration device which detects the patches included in the road footage, performs calibration on preset location information of each patch and pixel coordinate information of each patch in the road footage to generate a homography matrix.

Description

Calibration system and method {SYSTEM AND METHOD FOR PERFORMING CALIBRATION}

The present invention relates to a system and method for performing calibration.

The camera-based speed control system obtains the vehicle's location information through the camera, and detects the vehicle speed by calculating the time and the distance traveled. Since the speed control system can define the location information of the vehicle by matching the coordinates of the image collected by the camera and the actual physical location of the vehicle obtained by the distance sensor, the calibration process of the camera and the distance sensor is essential. .

On the other hand, the calibration was performed by selecting the approximate distance area from the collected image. However, the conventional method has a problem that an error occurs due to the interpolation technique using the interpolation technique in the case of an intermediate distance, and the accuracy is lowered. In addition, since the speed regulation system is installed on the road, the detection environment changes over time due to the external environment, and there is a problem that calibration must be performed again. In addition, the conventional method is complicated by the process of manually selecting the reference pixels in the image to perform the calibration, and requires a very large calibration board in the shape of a chessboard. There was.

Therefore, the present invention performs calibration by automatically determining the reference pixels in the image collected by the camera, and automatically updates the homography matrix generated by the calibration to perform adaptive calibration even when the detection environment changes. Provide technology.

The calibration system according to an embodiment of the present invention detects at least four patches disposed on a road, a camera for photographing a road image, and the patches included in the road image, preset position information of each patch, and And a calibration device that generates a homography matrix by calibrating pixel coordinate information of each patch in the road image.

The patches are respectively disposed at different preset positions on the road with respect to the camera.

The patches are circular shapes of the same size, each with a different internal pattern.

The calibration apparatus determines a first vanishing point by using a point at infinity on the physical space of the road image, and performs a second image processing on the road image to determine a second vanishing point. The homography matrix is corrected using the difference between the first vanishing point and the second vanishing point.

The calibration apparatus determines the horizon in the road image by projecting the infinite origin into the road image using the homography matrix, and determines the horizon as the first vanishing point.

The calibration apparatus performs image processing on the road image to identify a plurality of lanes existing in the road image, and determines an intersection point where the plurality of lanes meet as the second vanishing point.

The calibration apparatus compares the first vanishing point and the second vanishing point to determine a pitch angular displacement of the camera, and corrects the homography matrix using the pitch angular displacement.

According to another embodiment of the present invention, a calibration system detects at least four patches disposed on a road, distance sensors located on each patch, a camera photographing a road image, and the patches included in the road image. And a calibration device for obtaining location information of each patch measured by the distance sensors, and performing a calibration on the pixel information of each patch in the location information and the road image to generate a homography matrix.

The patches are respectively disposed at different preset positions on the road with respect to the camera.

The patches are rectangular in shape of equal size.

The patches each include a display panel for digitizing and displaying position information measured by a distance sensor corresponding to each patch, and the calibration device recognizes a numerical value included in the display panel to obtain the position information.

The calibration device determines a first vanishing point by using an infinite zero point in the physical space of the road image, and performs image processing on the road image to determine a second vanishing point, and between the first vanishing point and the second vanishing point. The homography matrix is corrected using the difference of.

The calibration apparatus determines the horizon in the road image by projecting the infinite origin into the road image using the homography matrix, and determines the horizon as the first vanishing point.

The calibration apparatus performs image processing on the road image to identify a plurality of lanes existing in the road image, and determines an intersection point where the plurality of lanes meet as the second vanishing point.

The calibration apparatus compares the first vanishing point and the second vanishing point to determine a pitch angular displacement of the camera, and corrects the homography matrix using the pitch angular displacement.

According to the present invention, calibration is performed by automatically determining the reference pixels, eliminating the need for a user operation for calibration and using a very large calibration board, which simplifies the calibration process and improves ease of use. have.

In addition, according to the present invention, the homography matrix is automatically updated in accordance with the change of detection environment, so that the performance of the speed regulation system using the present invention can be continuously maintained.

1 is a diagram illustrating a calibration process.
2 is a diagram illustrating a calibration system according to an embodiment of the present invention.
3 is a diagram illustrating exemplary patches in accordance with one embodiment of the present invention.
4 is a diagram illustrating how exemplary patches are disposed on a road according to one embodiment.
FIG. 5 is a diagram for describing a method of generating, by a calibration apparatus, a homography matrix through calibration using a road image, according to an exemplary embodiment.
FIG. 6 is a diagram for explaining a method of converting a road image using a generated homography matrix.
FIG. 7 is a diagram comparing a road image converted by a homography matrix requiring correction with a road image normally converted.
8 is a diagram for describing a method of calibrating a homography matrix by a calibration apparatus according to an exemplary embodiment.
9 is a view for explaining a calibration system according to another embodiment of the present invention.
10 is a diagram illustrating exemplary patches and distance sensors in accordance with another embodiment of the present invention.
11 is a diagram illustrating how exemplary patches are disposed on a road according to another embodiment.
FIG. 12 is a diagram for describing a method of generating, by a calibration device, a homography matrix through calibration using a road image, according to another exemplary embodiment.
FIG. 13 is a diagram for describing a method of calibrating a received road image by a calibration system.

DETAILED DESCRIPTION Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. As those skilled in the art would realize, the described embodiments may be modified in various different ways, all without departing from the spirit or scope of the present invention. In the drawings, parts irrelevant to the description are omitted in order to clearly describe the present invention, and like reference numerals designate like parts throughout the specification.

Throughout the specification, when a part is said to "include" a certain component, it means that it can further include other components, without excluding other components unless specifically stated otherwise.

In the present specification, "calibration" refers to a process of grasping geometric information, that is, a relationship between an image collected by a camera and a real physical space.

1 is a diagram illustrating a calibration process.

Referring to FIG. 1, the points x ₁ , x ₂ , x ₃ and x ₄ in the collected image 10 correspond to rays l ₁ , l ₂ , l ₃ and l ₄ passing through the focus c. Correspond to points X ₁ , X ₂ , X ₃ and X ₄ on the actual physical space 20. In this case, defined by the condition that the points on the real physical space 20 are on the same plane 30, the same plane as the points (x ₁ , x ₂ , x ₃ and x ₄ ) in the collected image 10. Points X ' ₁ , X' ₂ , X ' ₃ and X' ₄ on 30 may correspond. This correspondence process is called calibration, and when calibration is performed, the points X ₁ , x ₂ , x ₃ and x ₄ in the collected image 10 and the points X ' ₁ , X' ₂ on the same plane 30 are obtained. , A homography matrix 40 is created that represents the relationship of X ' ₃ and X' ₄ .

Hereinafter, a calibration system and method according to an exemplary embodiment of the present invention will be described with reference to the accompanying drawings.

2 is a diagram illustrating a calibration system according to one embodiment of the invention, FIG. 3 is a diagram illustrating exemplary patches in accordance with one embodiment of the present invention, and FIG. 4 is an exemplary patch according to one embodiment. And FIG. 5 is a diagram illustrating a method of generating a homography matrix through calibration using a road image according to an embodiment, and FIG. 6 is a diagram illustrating a method in which a calibration apparatus is arranged on a road. A diagram for describing a method of converting a road image using a homography matrix.

Referring to FIG. 2, the calibration system 100 includes patches 110, a camera 120, and a calibration device 130.

The patches 110 are composed of at least four, each having a circular shape of the same size having a different internal pattern.

For example, referring to FIG. 3, the first patch 111, the second patch 112, the third patch 113 and the fourth patch 114 may have concentric shapes having the same diameter. Internal patterns of the first patch 111, the second patch 112, the third patch 113, and the fourth patch 114 may be different from each other. Meanwhile, internal patterns of the first patch 111, the second patch 112, the third patch 113, and the fourth patch 114 shown in FIG. 3 are merely exemplary, and the first patch 111 and the first patch 111 may be formed. As long as the second patch 112, the third patch 113, and the fourth patch 114 may have different inner patterns, the shape of the specific inner pattern is not limited to any one.

In addition, the patches 110 are respectively disposed at different positions preset based on the camera 120 on the road.

For example, referring to FIG. 4, the first patch 111 is disposed 20m in the left direction and 20m in the upward direction based on the position of the camera 120, and the second patch 112 is disposed in the camera 120. 20 m in the left direction and 40 m in the upward direction based on the position of the third patch 113 is disposed 20 m in the right direction and 40 m in the upward direction based on the position of the camera 120, the fourth patch The 114 may be disposed 20 m in the right direction and 20 m in the upward direction based on the position of the camera 120.

In the above example, if the position of the camera 120 is (0m, 0m), the preset position information of the first patch 111 is (-20m, 20m), and the preset position of the second patch 112 The information is (-20m, 40m), the preset position information of the third patch 113 is (20m, 40m), and the preset position information of the fourth patch 114 may correspond to (20m, 20m). have.

The camera 120 photographs a road image.

In detail, the camera 120 captures an image of the road where the patches 110 are arranged as shown in FIG. 4. In this case, the camera 120 is a camera provided in the speed regulation system, and may be an apparatus for collecting an image of a road.

The calibration device 130 detects the patches 110 included in the road image.

Specifically, referring to FIG. 5, the calibration device 130 receives a road image from the camera 120 and performs an image processing technique on the received road image to detect patches 110 included in the road image. .

To this end, the calibration device 130 may pre-store an image of the patches 110 and detect the patches 110 in the road image using the pre-stored image. A method of detecting a specific image in an image using an image processing technique is well known and thus a detailed description thereof will be omitted.

The calibration device 130 determines the location information of the patches 110.

Specifically, referring to FIG. 5, when the calibration apparatus 130 detects the patches 110, the calibration apparatus 130 determines location information corresponding to each of the detected patches. To this end, the calibration device 130 may additionally store location information corresponding to each image, as well as an image of the patches 110.

For example, when the calibration device 130 detects the first patch 111, the second patch 112, the third patch 113, and the fourth patch 114 from the road image, the calibration apparatus 130 may store the stored location information. Based on the position information of the first patch 111 as (-20m, 20m), the position information of the second patch 112 as (-20m, 40m), and the position information of the third patch 113 as ( 20 m and 40 m), the position information of the fourth patch 114 may be determined as (20 m and 20 m), respectively.

The calibration device 130 generates a homography matrix by performing calibration on preset position information of each patch and pixel coordinate information of each patch in the road image.

Specifically, referring to FIG. 5, the calibration apparatus 130 determines pixel coordinate information of the patches 110 detected in the road image, and performs calibration on the position information and the pixel coordinate information of the patches 110, respectively. In this manner, a regular matrix is generated so that the position information of the patches 110 and the pixel coordinate information of the patches 110 correspond to each other. The generated regular matrix is called a homography matrix, and the process of generating a homography matrix corresponds to a calibration.

When the homography matrix is generated, the calibration device 130 may convert the received road image using the generated homography matrix to generate a converted image suitable for a speed control system such as a bird's eye view. For example, referring to FIG. 6, the calibration device 130 may convert the received road image into a bird's eye view using a homography matrix, and the speed control system equipped with the calibration device 130 may use a bird's eye view. It is possible to control whether the vehicle is overspeed.

The calibration system 100 locates the patches 110 at predefined locations on the road and automatically detects the patches 110 in the road image, so that, unlike the conventional method, a person manually changes the patches 110 in the road image. It is not necessary to select the position of). In addition, the patches 110 is configured in the form of a circular rather than the conventional chessboard calibration board, it can be made very small compared to the existing calibration board, there is an advantage that the arrangement is easy.

Therefore, according to the present invention, it is not necessary to use a very large calibration board without requiring a user operation for calibration, so that the calibration process can be simplified and the convenience for use can be improved.

7 is a diagram comparing a road image converted by a homography matrix requiring correction with a road image normally converted.

Referring to FIG. 7, even if a homography matrix is generated by performing calibration according to the initial position of the camera 120, an error may occur in the homography matrix according to a change in the position of the camera 120 due to an external environment. In detail, when the camera 120 is implemented in the speeding system, the camera 120 may be shaken in the vertical direction by external factors such as wind. In this case, the pitch angle of the camera 120 may change, and it is necessary to correct the homography matrix by reflecting this.

For example, when the road image received from the camera 120 is converted into a normal homography matrix, a correct bird's eye view may be generated as in the first converted image, but when the road image received from the camera 120 is converted into a homography matrix requiring correction, the second converted image may be generated. As described above, the bird's eye view may be generated in which the position of the vehicle and the relative position information of the lane are incorrect.

Accordingly, the calibration device 130 determines a first vanishing point using a point at infinity on the physical space of the road image, performs image processing on the road image, and determines a second vanishing point. The homography matrix is corrected using the difference between and the second vanishing point.

8 is a diagram for describing a method of calibrating a homography matrix by a calibration apparatus according to an embodiment.

Referring to FIG. 8, the calibration apparatus 130 determines a horizon from a road image by projecting an infinite origin into a road image using a homography matrix, and determines the horizon as a first vanishing point.

Specifically, the infinite origin refers to a virtual point where parallel lines meet in physical space. When the infinite origin existing on the road, which is a physical space, is projected on the road image using a homography matrix, the infinite origin corresponds to a vanishing point or a horizon where parallel lines meet in the road image. Therefore, the calibration device 130 may determine the horizon itself as the first vanishing point.

Meanwhile, in determining the horizon, the calibration device 130 does not calculate all the infinite origins in the physical space with the homography matrix, but connects the two horizontal ends to the road image by a line, and connects the connected lines to the horizon. You can also decide.

In addition, the calibration device 130 performs image processing on the road image to identify a plurality of lanes existing in the road image, and determines an intersection point where the plurality of lanes meet as a second vanishing point. In this case, the calibration device 130 may perform a filtering technique through tracking, such as a Kalman filter. A method of identifying linear components such as lanes and obtaining intersection points of the linear components by using an image processing technique with filtering is well known and thus detailed description thereof will be omitted.

Thereafter, the calibration device 130 compares the first vanishing point and the second vanishing point to determine the pitch angle displacement of the camera 120, and corrects the homography matrix using the pitch angle displacement.

Specifically, the homography matrix is composed of a product of an inner matrix and an outer matrix, which are defined in Equation 1 below.

In Equation 1, H ₀ is a homography matrix, K is an inner matrix, and S means an outer matrix.

According to Equation 1, the outer matrix may be represented by the product of the inverse matrix of the inner matrix and the homography matrix, and the matrix values of the outer matrix mean a column vector according to rotation transformation and a column vector according to parallel movement. This is defined as in Equation 2.

In Equation 2, t means a parallel shift component. In addition, r ₁ and r ₂ are rotation transformation components, and specifically, means rotation transformation about a two-dimensional axis of the physical system. Here, in order to recover all three-dimensional axes, the rotation component matrix of the three-dimensional axis can be obtained using the cross products of r ₁ and r ₂ . This is defined as in Equation 3.

In Equation 3, R _W means the rotational component matrix of the three-dimensional axis.

In addition, if T ^C _W that performs the axis rotation from the physical system to the camera is defined as Equation 4.

In addition, T ^C _W and R _W can be multiplied to obtain R _C , which is a rotation component on the camera. This is defined as in Equation 5.

Through the components of R _C , the pitch angle may be defined as shown in Equation 6, and the pitch angle displacement may be defined as shown in Equation 7.

는 제1 소실점과 제2 소실점의 y축 좌표의 차이이고,

는 보상될 피치 각도 변위를 의미한다.In Equation 7,

Is the difference between the y-axis coordinates of the first vanishing point and the second vanishing point,

Means the pitch angular displacement to be compensated for.

를 이용하여 피치 각도 보상 회전 행렬 R'을 정의하면, 수학식 8과 같다.Defined in Equation 7

If the pitch angle compensation rotation matrix R 'is defined using Equation (8).

In addition, a new pitch angle compensation rotation matrix R _new may be obtained by multiplying R 'and R _W , which is defined by Equation 9 below.

In addition, H 'can be obtained by multiplying an internal matrix and R _new , which is defined by Equation 10.

In addition, a corrected homography matrix can be obtained using the first column of H ', the second column of H', and the third column of the homography matrix, which are defined by Equation (11).

As described above, the calibration apparatus 130 may determine the pitch angular displacement of the first vanishing point and the second vanishing point through Equations 1 to 11, and correct the homography matrix using the determined pitch angular displacement.

The calibration device 130 may correct the homography matrix on a frame basis.

According to the present invention, the homography matrix is automatically updated as the detection environment changes, so that the performance of the speed regulation system using the present invention can be continuously maintained.

9 is a diagram illustrating a calibration system according to another embodiment of the present invention, FIG. 10 is a diagram illustrating exemplary patches and distance sensors according to another embodiment of the present invention, and FIG. FIG. 12 is a diagram illustrating a method of arranging exemplary patches on a road, and FIG. 12 is a diagram illustrating a method of generating a homography matrix through calibration using a road image according to another exemplary embodiment.

Referring to FIG. 9, the calibration system 200 includes patches 210, distance sensors 220, a camera 230, and a calibration device 240.

The patches 210 consist of at least four and have a rectangular shape of the same size.

For example, referring to FIG. 10, the first patch 211, the second patch 212, the third patch 213 and the fourth patch 214 may have a rectangular shape in the shape of a license plate, and all of them. It may be the same size.

Distance sensors 220 are located on each of the patches. In this case, the distance sensors 220 measure a distance from the camera 230 to each of the patches where the distance sensors are located.

For example, referring to FIG. 10, the first distance sensor 221 is located on the top of the first patch 211, the second distance sensor 222 is located on the top of the second patch 212, and the third distance sensor The 223 may be located on the top of the third patch 213, and the fourth distance sensor 224 may be located on the top of the fourth patch 214. In addition, the distance sensors 220 may be rotatably positioned on the patches 210 to measure a distance from the camera 230 to each of the patches where the distance sensors are located.

On the other hand, the patches 210 are each disposed in a different position previously set with respect to the camera 230 on the road, and each includes a display panel for digitizing and displaying the position information measured by the distance sensor corresponding to each patch do.

For example, referring to FIG. 11, the first patch 211 is disposed 20m in the left direction and 20m in the upper direction based on the position of the camera 230, and the second patch 212 is the camera 230. The third patch 213 is disposed at a distance of 20m in the left direction and 40m in the upward direction based on the position of the camera, and the third patch 213 is disposed at a distance of 10m in the right direction and 40m in the upper direction based on the position of the camera 230 and the fourth patch. When the 214 is disposed 20 m in the right direction and 20 m in the upward direction based on the position of the camera 230, the first distance sensor 221 uses the LED panel to make the (20, 20) a second distance. The sensor 222 uses the LED panel (20, 40), the third distance sensor 223 using the LED panel (10, 40), the fourth distance sensor 224 using the LED panel (20, 20) may be displayed. In this case, since the display panel of the patches in the left direction and the display panel of the patches in the right direction are configured differently based on the position of the camera 230, only numerical values may be displayed on the display panel without displaying a sign.

If the display panel of the patches in the left direction and the display panel of the patches in the right direction are not configured differently based on the position of the camera 230, the first distance sensor 221 uses the LED panel (−). 20, 20, the second distance sensor 222 using the LED panel (-20, 40), the third distance sensor 223 using the LED panel (10, 40), the fourth distance The sensor 224 may display 20, 20 using an LED panel.

The camera 230 photographs a road image.

Referring to FIG. 12, the calibration device 240 detects the patches 210 included in the road image, and obtains location information of each patch measured through the distance sensors 220.

In detail, the calibration apparatus 240 acquires position information of each patch 210 by recognizing a numerical value included in the display panel of the patches 210 using a numerical recognition algorithm such as a license plate recognition algorithm.

The calibration device 240 determines pixel coordinate information of the patches 210 detected in the road image, respectively. In this case, the calibration device 240 may determine the pixel coordinate information of the patches 210 based on the pixels of the lower centers of the patches 210, respectively.

The calibration device 240 generates a homography matrix by calibrating the pixel coordinate information of the patches in the location information and the road image.

In detail, the calibration device 240 generates a homography matrix that allows the position information of the patches 110 and the pixel information of the patches 110 to correspond to each other.

The calibration system 200 may automatically determine the location information of the patches 210 randomly located on the road through the distance sensors 220 so that a human may manually patch the patches 210 in the road image unlike the conventional method. The process of selecting a location is unnecessary. In addition, the patches 210 are configured in the form of a license plate rather than a conventional chessboard-shaped calibration board, it can be made very small compared to the existing calibration board, there is an advantage that it is easy to deploy.

Therefore, according to the present invention, it is not necessary to use a very large calibration board without requiring a user operation for calibration, so that the calibration process can be simplified and the convenience for use can be improved.

In addition, the calibration device 240 corrects the homography matrix in the same manner as described with reference to FIGS. 7 and 8.

FIG. 13 is a diagram illustrating a method of performing calibration on a received road image by a calibration system.

Referring to FIG. 13, the calibration system receives a road image through a camera (S100).

The calibration system detects at least four patches included in the received road image (S110).

The calibration system acquires location information corresponding to each patch (S120).

In this case, the calibration system may obtain location information corresponding to each patch by using the preset location information for each patch, and obtain the location information measured by the distance sensors located on the respective patches. Location information corresponding to the patches may be obtained.

The calibration system obtains pixel coordinate information of each patch in the road image (S130).

The calibration system performs calibration on the location information and the pixel coordinate information of each patch to generate a homography matrix that allows the location information and the pixel coordinate information of each patch to correspond (S140). In this case, the generated homography matrix corresponds to the homography matrix generated according to the initial position of the camera.

The calibration system determines the first vanishing point using the infinite origin in the physical space of the road image, performs image processing on the road image to determine the second vanishing point, and uses the difference between the first vanishing point and the second vanishing point. The homography matrix is corrected (S150).

In detail, the calibration system uses a homography matrix to project the infinite origin into the road image to determine the horizon in the road image, and to determine the horizon as the first vanishing point.

In addition, the calibration system performs image processing on the road image to identify a plurality of lanes existing in the road image, and determines an intersection point where the plurality of lanes meet as a second vanishing point.

Then, the calibration system compares the first vanishing point and the second vanishing point to determine the pitch angle displacement of the camera, and corrects the homography matrix using the pitch angle displacement.

According to the present invention, calibration is performed by automatically determining the reference pixels, eliminating the need for a user operation for calibration and using a very large calibration board, which simplifies the calibration process and improves ease of use. have.

In addition, according to the present invention, the homography matrix is automatically updated in accordance with the change of detection environment, so that the performance of the speed regulation system using the present invention can be continuously maintained.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements of those skilled in the art using the basic concepts of the present invention defined in the following claims are also provided. It belongs to the scope of rights.

Claims (15)

As a calibration system,
At least four patches placed on the roadway,
A camera that shoots road footage, and
And a calibration apparatus for detecting the patches included in the road image, and performing a calibration on preset position information of each patch and pixel coordinate information of each patch in the road image to generate a homography matrix.
The calibration device
A first vanishing point is determined using a point at infinity in the physical space of the road image, and a second vanishing point is determined by performing image processing on the road image. And calibrate the homography matrix using the difference between the second vanishing points. The method of claim 1,
The patches
And a calibration system each disposed at a different position preset on the road based on the camera. The method of claim 2,
The patches
Calibration system of the same size circular shape, each with a different internal pattern. delete The method of claim 1,
The calibration device
And a horizon in the road image by projecting the infinite origin into the road image using the homography matrix, and determining the horizon as the first vanishing point. The method of claim 1,
The calibration device
And performing image processing on the road image to identify a plurality of lanes existing in the road image, and to determine an intersection where the plurality of lanes meet as the second vanishing point. The method of claim 1,
The calibration device
A calibration system for comparing the first vanishing point and the second vanishing point to determine a pitch angular displacement of the camera, and correcting the homography matrix using the pitch angular displacement. As a calibration system,
At least four patches placed on the roadway,
Distance sensors located on each patch,
A camera that shoots road footage, and
Detects the patches included in the road image, obtains location information of each patch measured by the distance sensors, and performs calibration on the location information and pixel coordinate information of each patch in the road image. A calibration device for generating a matrix,
The calibration device
A first vanishing point is determined using an infinite original point on the physical space of the road image, an image processing is performed on the road image to determine a second vanishing point, and the difference between the first vanishing point and the second vanishing point is used. Calibrating the homography matrix. The method of claim 8,
The patches
And a calibration system arranged at different positions preset based on the camera on the road. The method of claim 9,
The patches
Calibration system of the same size rectangular shape. The method of claim 10,
The patches each include a display panel for digitizing and displaying the position information measured by the distance sensor corresponding to each patch,
And a calibration device configured to acquire the location information by recognizing a numerical value included in the display panel. delete The method of claim 8,
The calibration device
And a horizon in the road image by projecting the infinite origin into the road image using the homography matrix, and determining the horizon as the first vanishing point. The method of claim 8,
The calibration device
And performing image processing on the road image to identify a plurality of lanes existing in the road image, and to determine an intersection where the plurality of lanes meet as the second vanishing point. The method of claim 8,
The calibration device
A calibration system for comparing the first vanishing point and the second vanishing point to determine a pitch angular displacement of the camera, and correcting the homography matrix using the pitch angular displacement.

Images