WO2021000544A1

WO2021000544A1 - Calibration apparatus, and radar and camera joint calibration method and system

Info

Publication number: WO2021000544A1
Application number: PCT/CN2019/128584
Authority: WO
Inventors: 蔡樱贤; 李理; 欧敏辉; 柯志达; 陈卫强; 苏亮
Original assignee: 厦门金龙联合汽车工业有限公司
Priority date: 2019-07-04
Filing date: 2019-12-26
Publication date: 2021-01-07
Also published as: CN110310339B; CN110310339A

Abstract

Disclosed are a calibration apparatus, and a radar and camera joint calibration method and system. The method comprises: S1: providing a calibration apparatus, wherein the calibration apparatus comprises a camera calibration target and a radar calibration target, and the calibration apparatus is located at a calibration position; S2: acquiring, by means of a camera, pictures of the calibration apparatus in different states, and testing the calibration apparatus by means of a radar, wherein the different states comprise the camera calibration target and the radar calibration target being located in the same plane and the two jointly rotating around a first rotation shaft at different angles, and the camera calibration target rotating around a second rotation shaft at a different angle; and S3: calculating calibration data according to radar test data corresponding to the time when the camera acquires the pictures, and outputting the calibration data, wherein the calibration data comprises: a yaw angle, a pitch angle and a translation vector between the radar and the camera. According to the method, the deviation of a mounting angle is taken into consideration, so that the unification of information obtained by a radar and a camera can be ensured, and more accurate obstacle recognition and tracking can be realized.

Description

A calibration device and a combined calibration method and system of radar and camera

Technical field

The present invention relates to the technical field of calibration, in particular to a calibration device and a combined calibration method and system of radar and camera.

Background technique

Sensor fusion is a trend in the development of automobile safety. The combined use of cameras and radars can synthesize the characteristics of vision sensors and distance sensors, learn from each other, and improve the overall perception and safety performance of automobiles. Specifically, for example, vehicle forward collision warning.

In a multi-sensor sensing system, to realize the information fusion of each sensor, a unified coordinate system is required. The external parameters of the camera and millimeter wave radar are jointly calibrated, which directly affects the accuracy of vehicle obstacle detection. At present, there is little research on the joint calibration of binocular camera and millimeter wave radar at home and abroad. In the prior art, the external parameters between the millimeter wave radar and the camera are usually obtained by directly measuring the displacement deviation, and the angle deviation is generally ignored.

Summary of the invention

In view of the above problems, the present invention aims to provide a calibration device and a combined radar and camera calibration method and system, which convert the displacement deviation into the characteristic displacement deviation of the calibration device, which is more convenient to measure and obtain better measurement accuracy. The deviation of the installation angle is also considered to ensure that the information obtained by the radar and the camera can be unified, which can achieve more accurate obstacle recognition and tracking.

The specific plan is as follows:

A joint calibration method of radar and camera includes the following steps:

S1: Provide a calibration device, the calibration device includes a camera calibration board and a radar calibration board, so that the calibration device is in the calibration position;

S2: Collect pictures of the calibration device in different states through the camera, and at the same time detect the calibration device by radar. The different states include that the camera calibration board and the radar calibration board are both in the same plane and both rotate differently around the first rotation axis. The angle of and the camera calibration board rotates at different angles around the second rotation axis;

S3: Calculate and output calibration data according to the radar detection data corresponding to the moment when the camera collects the picture. The calibration data includes the yaw angle, the pitch angle and the translation vector between the radar and the camera.

Further, the reflectivity of the radar to the radar calibration board is different from the reflectivity of the radar to other components in the calibration device.

Further, the first rotation axis is perpendicular to the horizontal plane, and the second rotation axis is not parallel to the first rotation axis.

Further, the calculation process of the pitch angle is:

(1) According to the pictures taken by the camera in different states, the internal parameters of the camera are calculated by Zhang Zhengyou's calibration method. The internal parameters include the pixel ratio in the x direction, the pixel ratio in the y direction, the focal length and the distance between the center pixel coordinates of the picture and the origin pixel coordinates The number of horizontal and vertical pixels that differ;

(2) Select multiple pictures where the camera calibration board and radar calibration board in the picture are both in the same plane but rotated around the first axis of rotation at different angles, for each picture: calculate the rotation matrix and translation matrix of the picture Calculate the pitch angle and yaw angle of the camera coordinate system relative to the camera calibration board coordinate system according to the rotation matrix, and calculate the difference between the yaw angle and the rotation angle of the camera calibration board and the radar calibration board in the picture around the first rotation axis;

(3) The average value of the yaw angle of the selected multiple pictures and the rotation angle of the camera calibration board and the radar calibration board around the first rotation axis in the picture is used as the yaw angle of the camera installation;

(4) The average value of the elevation angle of the camera coordinate system of the selected multiple pictures relative to the camera calibration board coordinate system is used as the elevation angle of the camera installation.

Further, the yaw angle between the radar and the camera is the sum of the yaw angle installed by the radar and the yaw angle installed by the camera; the yaw angle installed by the radar is based on the sum of the data of the radar calibration board area in the calibration device detected by the radar. The relationship between the radar coordinate system and the radar calibration board coordinate system is calculated.

Further, the translation vector between the radar and the camera includes: lateral deviation, longitudinal deviation and height deviation;

The calculation process of the lateral deviation and the longitudinal deviation includes:

(1) Calculate the deviation between the coordinates of the camera calibration board and the radar calibration board according to the positional relationship between the camera calibration board and the radar calibration board when they are in the same plane;

(2) Set the camera calibration board and the radar calibration board in the calibration device to be in the same plane and the plane is parallel to the image plane of the camera, so that the camera and radar are away from the calibration device while collecting data at the same time;

(3) Extract multiple pictures including the radar calibration board from the pictures taken by the camera, and extract the radar detection data corresponding to the time the picture was taken. According to the relationship between the radar coordinate system and the camera coordinate system, project the radar detection data to the corresponding On the picture

(4) After calculating the longitudinal deviation according to the pixel coordinate difference between the median point of the radar detection data in each picture and the origin of the radar calibration board, the horizontal deviation is calculated according to the triangle similarity principle;

(5) Set the average value of the longitudinal deviation and the horizontal deviation of the multiple pictures as the longitudinal deviation and the horizontal deviation between the radar and the camera.

Further, the height deviation is calculated manually by a measuring tool, and the height deviation is the difference between the installation height of the camera and the installation height of the radar.

A radar and camera joint calibration system includes a calibration device, a camera, a radar, and a processing terminal. The processing terminal includes a processor, a memory, and a computer program stored in the memory and running on the processor. The system implements the steps of the method described in the embodiment of the present invention.

A calibration device, characterized in that it comprises a camera calibration board and a radar calibration board, the camera calibration board can rotate around a second rotation axis, the camera calibration board and the radar calibration board can be in the same plane and both They rotate together around the first rotation axis.

Further, the calibration device includes a bracket, an inner frame mounted on the bracket and capable of rotating, a radar calibration board fixedly mounted on the inner frame, and a radar calibration board mounted on the inner frame and capable of rotating around a direction that is not parallel to the direction of rotation of the inner frame Rotating camera calibration board.

The present invention adopts the above technical scheme and has beneficial effects:

(1) While calibrating the internal parameters of the camera, it is possible to jointly calibrate the external parameters between the radar and the camera, unify the radar coordinate system and the camera coordinate system, and improve the efficiency of the calibration process.

(2) Taking into account the millimeter-wave radar installation yaw angle and the camera installation yaw angle, which have always been ignored, the calibration effect is better.

(3) The three-dimensional transformation relationship between the radar coordinate system and the camera coordinate is obtained through the inverse solution of the coordinate relationship of the calibration board, which improves the problem that the angle relationship and horizontal displacement relationship between the two installations are difficult to directly measure, and the calibration effect obtained is more accurate .

(4) The method is simple and easy to operate, and the device is simple.

(5) During the calibration process, the fusion verification of the camera and the radar can be directly observed through the pictures, and the calibration effect is guaranteed.

Description of the drawings

FIG. 1 is a schematic flowchart of Embodiment 1 of the present invention.

Figure 2 shows a schematic diagram of the structure of the calibration plate in this embodiment.

Fig. 3 is a schematic diagram showing the relationship between the radar coordinate system and the metal plate coordinate system in this embodiment.

Figure 4 shows a schematic diagram of the relationship between the black and white grid coordinates and the metal plate coordinates in this embodiment.

Detailed ways

To further illustrate the embodiments, the present invention is provided with drawings. These drawings are a part of the disclosure of the present invention, which are mainly used to illustrate the embodiments, and can cooperate with the relevant description in the specification to explain the operation principle of the embodiments. With reference to these contents, those of ordinary skill in the art should be able to understand other possible implementations and advantages of the present invention.

The present invention will now be further described with reference to the drawings and specific embodiments.

Example one:

Referring to Fig. 1, the present invention provides a joint calibration method of radar and camera, which includes the following steps:

Step 1: Set up the calibration device.

The calibration device includes a bracket 1, an inner frame mounted on the bracket 1 and capable of rotating around the vertical direction 2, a metal plate 4 fixedly mounted on the inner frame 2 and a rotating direction perpendicular to the direction of rotation of the inner frame 2. Black and white grid 3.

Because the millimeter wave radar has different reflectivity for metal and non-metal, in order to better extract the data of the metal plate 4 area from the collected radar data, therefore, in this embodiment, the metal plate 4 is different from the metal plate 4 The other parts of the s are made of different materials, that is, the metal plate 4 is made of metal, and the other parts except the metal plate 4 in the calibration device are made of non-metallic materials. The two can be interchanged.

In this embodiment, if the support 1 is placed on the ground, the rotation axis of the inner frame 2 is vertical, and the rotation axis of the metal plate 4 is in the horizontal plane. The specific structure is shown in Figure 2. The bracket 1 is a rectangular frame structure, and the bottom includes two feet for stable placement on the ground. The frame is vertically arranged at the middle of the two horizontal frames of the rectangular frame for the inner frame 2 to rotate. The inner frame 2 also has a rectangular frame structure, and the middle positions of the upper and lower horizontal frames are respectively connected to the first rotary shaft 5, so that the inner frame 2 can rotate around the first rotary shaft 5. The metal plate 4 and the black and white grid plate 3 are both arranged in the inner frame 2, wherein the black and white grid plate 3 is arranged above, and the metal plate 4 is arranged below. The two vertical frames on the left and right of the inner frame 2 are used for vertical installation. The second rotating shaft 6 on which the black and white grid plate 3 rotates and the middle position of the blackboard divider are respectively connected to the second rotation shaft 6 so that the black and white grid plate 3 can rotate around the second rotation axis 6.

The metal plate 4 and the second rotating shaft 6 are arranged in the same plane, so that the metal plate 4 and the black-and-white grid plate 3 can be in the same plane and both rotate around the first rotating shaft 5 together.

In this embodiment, for the convenience of setting and calculation, the first rotation axis 5 and the second rotation axis 6 are arranged perpendicular to each other, that is, one is perpendicular to the horizontal plane, and the other is parallel to the horizontal plane. In other embodiments, both can be At other suitable angles.

The above-mentioned embodiment of the calibration device is only a preferred structure of the calibration device, and there are no restrictions on the shape, size, and connection position of each component in the calibration device.

Step 2: Set the calibration device at the calibration position in front of the vehicle.

In this embodiment, the calibration device is set in front of the vehicle at a position corresponding to the center of the vehicle, and the black-and-white grid 3 in the calibration device is located in the center of the camera screen and occupies roughly one-half of the screen size. The bracket of the calibration device 1 Parallel to the front of the vehicle.

Step 3: Take a picture of the calibration device through the camera, and detect the calibration device through the millimeter wave radar.

In this embodiment, three groups of pictures are taken by a camera fixedly arranged on the vehicle, and each group of pictures includes 5 pictures. among them:

In the first group of 5 pictures, the horizontal rotation angle of the inner frame 2 is 0°, and the vertical flip angle of the black and white grid 3 is 0°, ±20°, and ±45°, respectively.

In the 5 pictures of the second group, the horizontal rotation angle of the inner frame 2 is 45°, and the vertical flip angle of the black and white grid 3 is 0°, ±20°, and ±45°, respectively.

In the third group of 5 pictures, the horizontal rotation angle of the inner frame 2 is -45°, and the vertical flip angle of the black and white grid 3 is 0°, ±20°, and ±45°, respectively.

The horizontal rotation angle is the angle at which the inner frame 2 rotates relative to the support 1, and the up and down flip angle is the angle at which the black and white grid plate 3 rotates relative to the inner frame 2.

Step 4: Extract and filter out the data corresponding to the metal plate 4 in the millimeter wave radar information according to the detection data of the millimeter wave radar.

In this embodiment, according to the time stamp at the time of shooting, the first picture is extracted (the horizontal rotation angle of the inner frame 2 is 0°, and the vertical flip angle of the black and white grid 3 is 0°), and the sixth picture (inner The horizontal rotation angle of the frame 2 is 45°, the vertical flip angle of the black and white grid 3 is 0°) and the 11th picture (the horizontal rotation angle of the inner frame 2 is -45°, and the vertical flip angle of the black and white grid 3 is 0 For the radar detection data corresponding to the time of °), the radar detection data is filtered, leaving only the radar detection data of the metal plate 4 area.

Step 5: Calculate the yaw angle α of the millimeter wave radar installation.

As shown in Figure 3, the relationship between the millimeter wave radar coordinate system and the metal plate 4 coordinate system is:

Y _P =0

Among them, X _r -Y _r represents the millimeter wave radar coordinate system, X _p -Y _p represents the metal plate 4 coordinate system, l ₀ is the Y _r value when X _r is 0, and α is the yaw angle of the millimeter wave radar installation.

Through the data corresponding to the metal plate 4 detected by the millimeter wave radar, the installation yaw angle α of the millimeter wave radar is calculated.

Step 6: Calculate the internal parameters of the camera.

In this embodiment, according to the 3 groups of 15 pictures taken by the camera, the internal parameters of the camera are calculated using the Zhang Zhengyou calibration method: x-direction pixel ratio 1/dx, y-direction pixel ratio 1/dy, focal length f, the center pixel coordinates of the picture and The number of horizontal and vertical pixels u ₀ , v ₀ that differ between the pixel coordinates of the origin.

And calculate the rotation matrix R ₀ and translation matrix T ₀ of the first picture, the rotation matrix R ₁ and translation matrix T ₁ of the sixth picture, and the rotation matrix R ₂ and translation matrix T _{2 of} the 11th picture.

Step 7: Calculate the external parameters of the camera.

For the rotation matrices R ₀ , R ₁ , and R _{2 of the} above-mentioned pictures 1, 6 and 11, the pitch angle β ₀ , β ₁ , β ₂ and the yaw angle α _{0 of the} camera coordinate system relative to the black and white grid 3 coordinate system are obtained. , Α ₁ , α ₂ . Calculate yaw angle α ₀ and 0 ° deviation Δ _0, α ₁ and 45 ° deviation of Δ _1, and a deviation of -45 ° α ₂ Δ _2. Take the average value Δ of Δ ₀ , Δ ₁ , and Δ ₂ as the yaw angle of the camera installation. Take the average value β of β ₀ , β ₁ , and β ₂ as the pitch angle of the camera installation.

Step 8: Calculate the deviation between the coordinates of the black and white grid 3 and the metal plate 4 according to the positional relationship between the black and white grid 3 and the metal plate 4.

X _W ＝X _P +a

Z _P =0

Y _W ＝bZ _P ＝b

Z _W ＝Y _P ＝0

Among them, X _W -Y _W is the black and white grid 3 coordinate system (ie the camera coordinate system), X _p -Y _p is the metal plate 4 coordinate system, and a is the lateral deviation between the black and white grid 3 coordinates and the metal plate 4 coordinates , B is the longitudinal deviation between the coordinates of the black and white grid plate 3 and the metal plate 4.

As shown in Fig. 4, in this embodiment, the origin of the coordinates of the black and white grid 3 is the first corner point in the upper left corner, and the origin of the coordinates of the metal plate 4 is the center of the metal plate 4. Measure the horizontal deviation a and the vertical deviation sum b between the coordinates of the black and white grid plate 3 and the metal plate 4 with a ruler.

Step 9: Obtain the lateral deviation lx and the longitudinal deviation ly of the millimeter wave radar and camera installation.

The horizontal rotation angle of the inner frame 2 of the calibration device is set to 0°, so that the vehicle moves slowly and straight back, and the camera and the millimeter wave radar collect data at the same time during the backward process.

Extract the synchronization frame data of 3-5 cameras and millimeter wave radar after the metal plate 4 also enters the camera screen, and project the radar detection data to the picture taken by the camera according to the relationship between the millimeter wave radar coordinates and the camera coordinate system :

Among them, the yaw angle α is obtained in step 5, 1/dx, 1/dy, f, u ₀ , and v _{0 are} obtained in step 6, R and T are selected from R ₀ and T _{0 in} step 6, and a and b are obtained from steps Eight obtained.

Accordingly, according to the pixel coordinate difference between the intermediate value point of the millimeter wave radar detection data in the picture and the origin of the metal plate 4, the longitudinal deviation l _{y is obtained} , and then the horizontal deviation l _{x is} obtained according to the triangle similarity principle.

Step 10: Measure the height installation deviation l _z .

In this embodiment, a ruler is used to measure the installation height h _{1 of the} millimeter wave radar, the camera installation height h ₂ , and the height installation deviation l _z =h ₂ -h ₁ .

Step 11: Output calibration data.

The conversion of the millimeter wave radar coordinate system to the camera coordinate system is a rigid body transformation, which can be regarded as the millimeter wave radar first has a translation transformation relative to the camera, and then a rotation transformation. Among them, the installation pitch and roll angles of the radar are ignored, and the camera is installed The roll angle is ignored. Then the yaw angle between the millimeter wave radar and the camera = α + Δ, the pitch angle is β, and the roll angle is 0, which are obtained by steps 5 and 7 respectively. The translation vector (l _x , l _y , l _z ) is obtained by step 9 and step 10 respectively. Take the yaw angle α+Δ, pitch angle β, roll angle 0, and translation vector (l _x , l _y , l _z ) between the millimeter wave radar and the camera as the data output of the millimeter wave radar and camera joint calibration. Used by the vehicle's subsequent perception module.

The method in the first embodiment of the present invention can be used in any intelligent driving system that uses both radar and camera sensors, such as AEBS.

This embodiment has the following improvements:

(1) Improved based on Zhang Zhengyou's calibration method. By fixing the angles of three calibration devices, while calibrating the internal parameters of the camera, the pitch angle and yaw angle of the camera installation are obtained.

(2) A calibration device is designed, and a strong reflection wave metal plate 4 is fixed on the calibration device at the same time. While calibrating the internal parameters of the camera, it can calibrate the yaw angle of the millimeter wave radar.

(3) Through the inverse solution of the coordinate relationship of the calibration device, the three-dimensional transformation relationship between the radar coordinate system and the camera coordinate is obtained, which improves the problem that the angle relationship and the horizontal displacement relationship between the two installations are difficult to directly measure.

It has the following advantages:

(3) Through the inverse solution of the coordinate relationship of the calibration device, the three-dimensional transformation relationship between the radar coordinate system and the camera coordinate is obtained, which improves the problem that the angle relationship and horizontal displacement relationship between the two installations are difficult to directly measure, and the obtained calibration effect is more accurate .

(4) The method is simple and easy to operate, and the device is simple.

(5) During the calibration process, the fusion verification of the two sensor data can be directly observed through the pictures, and the calibration effect is guaranteed.

Embodiment two:

The present invention also provides a radar and camera joint calibration system, including the calibration device as described in the first embodiment, a camera, a radar, and a processing terminal. The processing terminal includes a memory, a processor, and a bus. The memory stores at least A piece of program, the system implements the radar and camera joint calibration method as described in the first embodiment of the present invention.

The processor includes one or more processing cores, the processor is connected to the memory through a bus, and the memory is used to store program instructions. When the processor executes the program instructions in the memory, the radar and camera joint calibration described in the first embodiment of the present invention are implemented. method.

Further, as an executable solution, the so-called processor may be a central processing unit (Central Processing Unit, CPU), or other general-purpose processors, digital signal processors (Digital Signal Processors, DSPs), and application-specific integrated circuits (Central Processing Unit, CPU). Application Specific Integrated Circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc. The processor is the control center of the radar and camera joint calibration system, which uses various interfaces and lines to connect the entire radar and camera All parts of the joint calibration system.

The memory may be used to store the computer program and/or module, and the processor realizes the radar by running or executing the computer program and/or module stored in the memory and calling the data stored in the memory. Various functions of the calibration system in conjunction with the camera. The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system and an application program required by at least one function; the data storage area may store data created according to the use of the mobile phone, etc. In addition, the memory may include high-speed random access memory, and may also include non-volatile memory, such as hard disk, memory, plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card , Flash Card, at least one magnetic disk storage device, flash memory device, or other volatile solid-state storage device.

Although the present invention has been specifically shown and described in conjunction with the preferred embodiments, those skilled in the art should understand that the present invention can be modified in form and detail without departing from the spirit and scope of the present invention as defined by the appended claims. Various changes are within the protection scope of the present invention.

Claims

A joint calibration method of radar and camera is characterized in that it comprises the following steps:

S1: Provide a calibration device, the calibration device includes a camera calibration board and a radar calibration board, so that the calibration device is in the calibration position;

S2: Collect pictures of the calibration device in different states through the camera, and at the same time detect the calibration device by radar. The different states include that the camera calibration board and the radar calibration board are both in the same plane and both rotate differently around the first rotation axis. The angle of and the camera calibration board rotates at different angles around the second rotation axis;

S3: Calculate and output calibration data according to the radar detection data corresponding to the moment when the camera collects the picture. The calibration data includes the yaw angle, the pitch angle and the translation vector between the radar and the camera.
The method according to claim 1, wherein the reflectivity of the radar to the radar calibration plate is different from the reflectivity of the radar to other components in the calibration device.
The method according to claim 1, wherein the first rotation axis is perpendicular to the horizontal plane, and the second rotation axis is not parallel to the first rotation axis.
The method according to claim 1, wherein the calculation process of the pitch angle is:

(1) According to the pictures taken by the camera in different states, the internal parameters of the camera are calculated by Zhang Zhengyou's calibration method. The internal parameters include the pixel ratio in the x direction, the pixel ratio in the y direction, the focal length and the distance between the center pixel coordinates of the picture and the origin pixel coordinates The number of horizontal and vertical pixels that differ;

(2) Select multiple pictures where the camera calibration board and radar calibration board in the picture are both in the same plane but rotated around the first axis of rotation at different angles, for each picture: calculate the rotation matrix and translation matrix of the picture Calculate the pitch angle and yaw angle of the camera coordinate system relative to the camera calibration board coordinate system according to the rotation matrix, and calculate the difference between the yaw angle and the rotation angle of the camera calibration board and the radar calibration board in the picture around the first rotation axis;

(3) The average value of the yaw angle of the selected multiple pictures and the rotation angle of the camera calibration board and the radar calibration board around the first rotation axis in the picture is used as the yaw angle of the camera installation;

(4) The average value of the elevation angle of the camera coordinate system of the selected multiple pictures relative to the camera calibration board coordinate system is used as the elevation angle of the camera installation.
The method according to claim 4, wherein the yaw angle between the radar and the camera is the sum of the yaw angle installed by the radar and the yaw angle installed by the camera; the yaw angle installed by the radar is based on the yaw angle detected by the radar The data of the radar calibration board area in the calibration device and the relationship between the radar coordinate system and the radar calibration board coordinate system are calculated.
The method according to claim 1, wherein the translation vector between the radar and the camera includes: lateral deviation, longitudinal deviation and height deviation;

The calculation process of the lateral deviation and the longitudinal deviation includes:

(1) Calculate the deviation between the coordinates of the camera calibration board and the radar calibration board according to the positional relationship between the camera calibration board and the radar calibration board when they are in the same plane;

(2) Set the camera calibration board and the radar calibration board in the calibration device to be in the same plane and the plane is parallel to the image plane of the camera, so that the camera and radar are away from the calibration device while collecting data at the same time;

(3) Extract multiple pictures including the radar calibration board from the pictures taken by the camera, and extract the radar detection data corresponding to the time the picture was taken. According to the relationship between the radar coordinate system and the camera coordinate system, project the radar detection data to the corresponding On the picture

(4) After calculating the longitudinal deviation according to the pixel coordinate difference between the median point of the radar detection data in each picture and the origin of the radar calibration board, the horizontal deviation is calculated according to the triangle similarity principle;

(5) Set the average value of the longitudinal deviation and the horizontal deviation of the multiple pictures as the longitudinal deviation and the horizontal deviation between the radar and the camera.
The method according to claim 6, characterized in that the height deviation is manually measured by a measuring tool, and the height deviation is the difference between the installation height of the camera and the installation height of the radar.
A combined radar and camera calibration system, which is characterized in that it includes a calibration device, a camera, a radar, and a processing terminal. The processing terminal includes a processor, a memory, and a device stored in the memory and running on the processor. A computer program that implements the steps of the method as claimed in any one of claims 1-7.
A calibration device, characterized in that it comprises a camera calibration board and a radar calibration board, the camera calibration board can rotate around a second rotation axis, the camera calibration board and the radar calibration board can be in the same plane and both They rotate together around the first rotation axis.
The calibration device according to claim 9, characterized in that: the calibration device comprises a bracket, a rotatable inner frame installed on the bracket, a radar calibration board fixedly installed on the inner frame, and a radar calibration board installed on the inner frame and capable of A camera calibration board that rotates around a direction that is not parallel to the rotation direction of the inner frame.