A camera extrinsic parameter calibration method, device and system

Abstract

The application discloses a camera external parameter calibration method, device and system, obtains the internal parameter matrix and distortion coefficient of a camera calibrated by adopting Zhang Zhengyou camera calibration method, obtains the lateral distance and the height from the ground of the camera relative to the vehicle collected by an external parameter calibration auxiliary support device, and calibrates the rotation angle of the camera and the distance of the camera relative to the front bumper based on the internal parameter matrix, the distortion coefficient, the lateral distance and the height from the ground, so as to complete the calibration of the external parameter of the camera. The application can calibrate the external parameter of the camera before the vehicle drives based on the external parameter calibration auxiliary support device, and manual calibration of the external parameter of the camera is not needed during the driving of the vehicle, so that the calibration of the external parameter of the camera is more convenient, real-time sensing of the target position on the camera is realized, the target can be visually sensed, and the development and test of the automatic driving function are facilitated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of automatic driving, and more particularly to a camera external parameter calibration method, device and system. BACKGROUND

[0002] In the field of automatic driving, a single sensor cannot meet the accuracy requirements of vehicle perception of the surrounding environment, and multi-sensor fusion has become a technical trend. Monocular camera visualizes the multiple targets perceived by multiple sensors (such as perception cameras, millimeter wave radars, laser radars, etc.) to intuitively understand the perception performance of each sensor. At this time, the intrinsic matrix, distortion coefficient and external parameter of the monocular camera need to be calibrated first, and then the coordinate system is converted according to the target position information returned by the sensor, so that the target can be intuitively marked on the camera image.

[0003] At present, when calibrating the external parameter, it is usually manually adjusted according to the actual picture collected by the camera based on experience, which often needs to be adjusted several times, and the adjustment is made during vehicle driving. Therefore, the external parameter calibration is very inconvenient. SUMMARY

[0004] Therefore, the present application discloses a camera external parameter calibration method, device and system to calibrate the camera external parameter before the vehicle drives, without manually calibrating the camera external parameter during the vehicle driving, so that the camera external parameter calibration is more convenient, the target position can be perceived in real time on the camera, and the target can be visually perceived, thereby providing convenience for the development and testing of automatic driving functions.

[0005] A camera external parameter calibration method, comprising:

[0006] obtaining the intrinsic matrix and distortion coefficient of the camera calibrated by Zhang Zhengyou camera calibration method;

[0007] obtaining the lateral distance and height from the ground of the camera relative to the vehicle collected by an external parameter calibration auxiliary support device;

[0008] based on the intrinsic matrix, the distortion coefficient, the lateral distance and the height from the ground, calibrating the camera rotation angle and the distance of the camera relative to the front bumper, and completing the calibration of the camera external parameter.

[0009] Optionally, the external parameter calibration auxiliary support device comprises a base support and a sliding support, and the sliding support is vertically fixedly installed on the base support;

[0010] a sliding plate with a fixed length is arranged on the cross bar of the sliding support, and a laser is installed at the center of the sliding plate;

[0011] The crossbar is capable of moving along the sliding support, and the sliding plate is capable of moving along the crossbar.

[0012] Optionally, the base support is further provided with a link, which is used to lengthen two feet of the base support.

[0013] Optionally, the lateral distance and the height from the ground are acquired by the external parameter calibration auxiliary support device, and the acquisition includes:

[0014] When the external parameter calibration auxiliary support device is placed at a preset position so that the center of the laser beam emitted by the laser falls on the camera lens position, the height of the laser from the ground and the center offset of the laser relative to the crossbar are acquired.

[0015] The height of the laser from the ground is determined as the height from the ground, and the center offset is determined as the lateral distance.

[0016] The preset position is that the base support is aligned with the front wheels of the vehicle, the surface formed by the sliding support is in contact with the front bumper of the vehicle, the base support is close to the two front wheels of the vehicle, the direction of the laser beam emitted by the laser is consistent with the direction of the feet of the base support, and the laser beam and the camera lens are at the same height.

[0017] Optionally, a recess is arranged at the middle position of the sliding plate, and the laser is installed at the position where the recess is located.

[0018] Optionally, based on the intrinsic matrix, the distortion coefficient, the lateral distance and the height from the ground, the rotation angle of the camera and the distance of the camera relative to the front bumper are calibrated, and the calibration of the external parameter of the camera is completed, and the calibration includes:

[0019] An external parameter calibration image containing the entire sliding plate is acquired by the camera, wherein the sliding plate is a mechanism for placing the laser in the external parameter calibration auxiliary support device.

[0020] Two mark points are marked at both ends of the sliding plate image on the external parameter calibration image.

[0021] Based on the intrinsic matrix, the distortion coefficient, the lateral distance and the height from the ground, the value of the rotation angle of the camera is continuously changed, wherein the initial value of the rotation angle of the camera is 0, and the rotation angle of the camera includes a yaw angle, a pitch angle and a roll angle.

[0022] The two mark points on the external parameter calibration image are dynamically moved to the half-height position of the image, and the two mark points are symmetrical along the longitudinal center axis of the external parameter calibration image.

[0023] record the opposite number of the target camera rotation angle when two of the mark points are located at the half-height positions of the image and two of the mark points are symmetrical along the longitudinal center axis of the extrinsic parameter calibration image, and determine the opposite number as the camera rotation angle calibration result;

[0024] obtain the distance of the camera relative to the front bumper based on the pixel distance of two of the mark points on the extrinsic parameter calibration image and the length of the slide plate.

[0025] Optionally, the continuously changing of the value of the camera rotation angle based on the intrinsic matrix, the distortion coefficient, the lateral distance amount and the distance-ground height amount comprises:

[0026] the continuously changing of the value of the camera rotation angle based on the intrinsic matrix, the distortion coefficient, the lateral distance amount and the distance-ground height amount by dragging the slide bar through the slide bar callback function.

[0027] Optionally, the changing of the size of the yaw angle causes the mark points to move along the image transverse axis on the extrinsic parameter calibration image, the changing of the size of the pitch angle causes the mark points to move along the image longitudinal axis on the extrinsic parameter calibration image, and the changing of the size of the roll angle causes two of the mark points to rotate around the preset center point of the extrinsic parameter calibration image.

[0028] Optionally, the obtaining of the distance of the camera relative to the front bumper based on the pixel distance of two of the mark points on the extrinsic parameter calibration image and the length of the slide plate comprises:

[0029] obtaining the distance of the camera relative to the front bumper according to the imaging principle and the following formula:

[0030] d=f*L / p;

[0031] wherein, d is the distance of the camera relative to the front bumper, f is the focal length of the camera, L is the length of the slide plate, and p is the pixel distance of two of the mark points on the extrinsic parameter calibration image.

[0032] A camera extrinsic parameter calibration device, comprising:

[0033] a first acquisition unit configured to acquire an intrinsic matrix and a distortion coefficient of a camera calibrated by Zhang Zhengyou camera calibration method;

[0034] a second acquisition unit configured to acquire a lateral distance amount and a distance-ground height amount of the camera relative to a vehicle collected by an extrinsic parameter calibration auxiliary support device;

[0035] An external parameter calibration unit is configured to calibrate a camera rotation angle and a distance of the camera relative to a front bumper based on the intrinsic matrix, the distortion coefficient, the lateral distance amount and the distance-ground height amount, and complete calibration of the camera external parameter.

[0036] A camera external parameter calibration system includes the external parameter calibration auxiliary support device and the camera external parameter calibration device.

[0037] According to the technical solution, the intrinsic matrix and the distortion coefficient of the camera calibrated by the Zhang Zhengyou camera calibration method are obtained, the lateral distance amount and the distance-ground height amount of the camera relative to the vehicle are obtained by the external parameter calibration auxiliary support device, the camera rotation angle and the distance of the camera relative to the front bumper are calibrated based on the intrinsic matrix, the distortion coefficient, the lateral distance amount and the distance-ground height amount, and the calibration of the camera external parameter is completed. The camera external parameter calibration can be performed before the vehicle drives based on the external parameter calibration auxiliary support device, and manual calibration of the camera external parameter is not required during the driving of the vehicle, so that the camera external parameter calibration is more convenient, real-time sensing of the target position on the camera is realized, the target can be visually sensed, and the development and testing of the automatic driving function are facilitated. BRIEF DESCRIPTION OF DRAWINGS

[0038] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed in the embodiments or the prior art description will be briefly introduced. Obviously, the drawings in the following description are only embodiments of the present application, and other drawings can be obtained by the disclosed drawings without creative labor for those skilled in the art.

[0039] Figure 1 A camera external parameter calibration method flowchart disclosed by the embodiment of the present application;

[0040] Figure 2 A structural schematic diagram of an external parameter calibration auxiliary support device disclosed by the embodiment of the present application;

[0041] Figure 3 A front view of an external parameter calibration auxiliary support device disclosed by the embodiment of the present application;

[0042] Figure 4 A top view of an external parameter calibration auxiliary support device disclosed by the embodiment of the present application;

[0043] Figure 5 A relative position schematic diagram of an external parameter calibration auxiliary support device and a vehicle disclosed by the embodiment of the present application;

[0044] Figure 6 A relative position diagram of an external parameter calibration auxiliary support device and a camera is disclosed for an embodiment of the present application.

[0045] Figure 7 A method flowchart for calibrating an external parameter of a camera is disclosed for an embodiment of the present application, which calibrates a rotation angle of the camera and a distance of the camera relative to a front bumper based on an intrinsic matrix, distortion coefficients, a lateral distance amount and a distance-ground height amount.

[0046] Figure 8 An external parameter calibration image schematic diagram collected by a camera and containing an entire skateboard is disclosed for an embodiment of the present application.

[0047] Figure 9 A structure schematic diagram of a camera external parameter calibration device is disclosed for an embodiment of the present application. DETAILED DESCRIPTION

[0048] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative labor fall within the scope of protection of the present application.

[0049] The embodiments of the present application disclose a camera external parameter calibration method, device and system, which acquire an intrinsic matrix and distortion coefficients of a camera calibrated by Zhang Zhengyou camera calibration method, acquire a lateral distance amount and a distance-ground height amount of the camera relative to a vehicle collected by an external parameter calibration auxiliary support device, calibrate a rotation angle of the camera and a distance of the camera relative to a front bumper based on the intrinsic matrix, the distortion coefficients, the lateral distance amount and the distance-ground height amount, and complete calibration of the external parameter of the camera. The present application can calibrate the external parameter of the camera before the vehicle drives based on the external parameter calibration auxiliary support device, without manually calibrating the external parameter of the camera during the driving of the vehicle, so that the calibration of the external parameter of the camera is more convenient, real-time perception of a target position on the camera is realized, and the target can be visually perceived, thereby providing convenience for the development and testing of automatic driving functions.

[0050] Referring to Figure 1 The camera external parameter calibration method flowchart disclosed by the embodiments of the present application comprises:

[0051] In step S101, the intrinsic matrix and distortion coefficients of the camera calibrated by Zhang Zhengyou camera calibration method are acquired.

[0052] Wherein, the intrinsic matrix includes 4 parameters, which are respectively: the pixel length of the focal length in the x-axis direction and the y-axis direction in the vehicle coordinate system, and the pixel coordinates of the image center in the x-axis direction and the y-axis direction in the vehicle coordinate system.

[0053] In practical application, a monocular camera is prepared, the focal length of the monocular camera is adjusted and kept unchanged, and a fixed-focus camera can be preferably selected, and the image resolution output by the fixed-focus camera is relatively high. A black and white chessboard is prepared, the chessboard has consistent square grids, the number of inner corner points is as large as possible, generally more than 3 (high) x 5 (wide), and the board surface is flat; the calibration board in the calibration image is free of glare, reflected light or other obstacle shielding; 15-20 images of each posture containing a complete chessboard are shot, the shooting angle is preferably 30-75 degrees to ensure the image quality while increasing the distinguishability, and the image is clear and free of ghosting; by using the camera calibration method of Zhang Zhengyou, the image and the number of inner corner points of the chessboard are input into the camera intrinsic parameter calibration program to obtain the intrinsic matrix and the distortion coefficient of the camera.

[0054] In step S102, the lateral distance amount and the distance from the ground height amount of the camera relative to the vehicle collected by the external parameter calibration auxiliary support device are acquired.

[0055] In step S103, the camera rotation angle and the distance of the camera relative to the front bumper are calibrated based on the intrinsic matrix, the distortion coefficient, the lateral distance amount and the distance from the ground height amount, and the calibration of the camera external parameter is completed.

[0056] Wherein, the camera external parameter includes: the camera rotation angle and the distance of the camera relative to the front bumper, the camera rotation angle includes: the yaw angle, the pitch angle and the roll angle, and the distance of the camera relative to the front bumper includes: three relative distances of the camera optical center relative to the vehicle.

[0057] As can be seen from the above, the present application discloses a camera external parameter calibration method, the intrinsic matrix and the distortion coefficient of the camera calibrated by using the camera calibration method of Zhang Zhengyou are acquired, the lateral distance amount and the distance from the ground height amount of the camera relative to the vehicle collected by the external parameter calibration auxiliary support device are acquired, the camera rotation angle and the distance of the camera relative to the front bumper are calibrated based on the intrinsic matrix, the distortion coefficient, the lateral distance amount and the distance from the ground height amount, and the calibration of the camera external parameter is completed. The present application is based on the external parameter calibration auxiliary support device, the camera external parameter calibration can be performed before the vehicle drives, manual calibration of the camera external parameter is not needed during the driving of the vehicle, so that the camera external parameter calibration is more convenient, real-time perception of the target position on the camera is realized, the target can be visually perceived, and the development and testing of the automatic driving function are facilitated.

[0058] It needs to be particularly pointed out that the external parameter calibration auxiliary support device used in step S102 is a convenient-to-disassemble support structure proposed by the application for camera external parameter calibration.

[0059] Therefore, in order to further optimize the above-mentioned embodiment, referring to Figure 2 The external parameter calibration auxiliary support device disclosed by the embodiment of the application comprises a base support 1 and a sliding support 2, and the sliding support 2 is vertically fixedly installed on the base support 1.

[0060] Combining the front view of the external parameter calibration auxiliary support device shown in Figure 3 and the top view of the external parameter calibration auxiliary support device shown in Figure 4 It can be known that a sliding plate 4 with a fixed length is arranged on the cross rod 3 of the sliding support 2, and a laser (not shown in the figure) is installed at the center of the sliding plate 4. Figure 2

[0061] In actual application, the middle position of the sliding plate 4 is provided with a groove, and the laser can be installed at the position of the groove.

[0062] In the external parameter calibration auxiliary support device, the cross rod 3 can move along the sliding support 2, and the sliding plate 4 can move along the cross rod 3.

[0063] In order to further optimize the above-mentioned embodiment, the base support 1 is further provided with a link 5, and the link 5 is used to lengthen the two feet of the base support 1.

[0064] When the external parameter calibration auxiliary support device is used to collect the lateral distance amount of the camera relative to the vehicle and the height amount of the camera relative to the ground, referring to Figure 5 and Figure 6 The base support 1 is aligned with the front wheels of the vehicle, the two feet of the base support 1 are lengthened through the link 5, the surface formed by the sliding support 2 is in contact with the front bumper of the vehicle, and the feet of the base support 1 can be close to the two front wheels. The laser 7 is installed at the middle groove position of the sliding plate 4 on the cross rod 3, so that the direction of the light beam is consistent with the direction of the feet of the base support 1. The laser 7 is turned on, the cross rod 3 is moved up and down until the laser beam 8 is at the same height as the camera lens 6, and then the cross rod 3 is fixed.

[0065] After the cross rod 3 is fixed, the sliding plate 4 on the cross rod 3 is moved left and right, so that the center of the laser beam 8 falls on the position of the camera lens 6, and then the sliding plate 4 is fixed. The height of the laser 7 from the ground and the offset amount relative to the center of the cross rod 3 are measured, and two distance amounts of the camera relative to the vehicle are directly obtained, that is, the lateral distance amount of the camera relative to the vehicle and the height amount of the camera relative to the ground.

[0066] Based on this, step S102 can specifically comprise:

[0067] ​When the external parameter calibration auxiliary support device is placed at the preset position so that the center of the laser beam emitted by the laser falls at the camera lens position, the height of the laser from the ground and the center offset of the laser relative to the crossbar are obtained;

[0068] The height of the laser from the ground is determined as the ground distance, and the center offset is determined as the lateral distance.

[0069] The preset position is that the base support is aligned with the front wheels of the vehicle, the surface formed by the sliding support is in contact with the front bumper of the vehicle, the base support is close to the two front wheels of the vehicle, the direction of the laser beam emitted by the laser is consistent with the direction of the foot of the base support, and the laser beam and the camera lens are at the same height.

[0070] To further optimize the above embodiment, referring to Figure 7 The embodiment of the application discloses a method for calibrating the external parameters of a camera based on an intrinsic matrix, distortion coefficients, a lateral distance and a ground distance, and the method comprises the steps of:

[0071] In step S201, an external parameter calibration image containing a whole slide plate is acquired by a camera.

[0072] As shown in Figure 2 The slide plate is a mechanism for placing a laser in an external parameter calibration auxiliary support device.

[0073] In step S202, two mark points are marked at both ends of a slide plate image on the external parameter calibration image.

[0074] As shown in Figure 8 The embodiment of the application discloses an external parameter calibration image containing a whole slide plate acquired by a camera, two mark points are marked at both ends of a slide plate image on the external parameter calibration image, and the two mark points are mark point 1 and mark point 2.

[0075] In step S203, the value of the rotation angle of the camera is continuously changed based on the intrinsic matrix, distortion coefficients, the lateral distance and the ground distance.

[0076] The initial value of the rotation angle of the camera is 0, and the unit is rad. The rotation angle of the camera comprises a yaw angle, a pitch angle and a roll angle.

[0077] Specifically, based on the internal reference matrix, distortion coefficient, transverse distance and distance from the ground height, the two marker points (marker point 1 and marker point 2) on the extrinsic calibration image are dynamically moved to the half-height position of the image and the two marker points are symmetrical along the longitudinal center axis of the extrinsic calibration image by continuously changing the value of the camera rotation angle by dragging the slider bar through the slider bar callback function.

[0078] The specific process is that the initial values of the yaw angle, the pitch angle and the roll angle of the camera are all 0 rad, and the sizes of the three rotation angles are dynamically changed by dragging the slider bar through the slider bar callback function. Changing the size of the yaw angle (range: -0.4 to 0.4 rad), the marker point moves along the image horizontal axis (i.e. the x-axis in the figure) on the extrinsic calibration image; changing the size of the pitch angle (range: -0.4 to 0.4 rad), the marker point moves along the image vertical axis (i.e. the y-axis in the figure) on the extrinsic calibration image; changing the size of the roll angle (range: -0.1 to 0.1 rad), the two marker points rotate around the preset center point of the extrinsic calibration image. Figure 8 Figure 8

[0079] Step S204, dynamically moving the two marker points on the extrinsic calibration image to the half-height position of the image and the two marker points being symmetrical along the longitudinal center axis of the extrinsic calibration image.

[0080] Step S205, recording the opposite number of the target camera rotation angle when the two marker points are located at the half-height position of the image and the two marker points are symmetrical along the longitudinal center axis of the extrinsic calibration image, and determining the opposite number as the camera rotation angle calibration result.

[0081] Step S206, based on the pixel distance of the two marker points on the extrinsic calibration image and the length of the slide plate, obtaining the distance of the camera relative to the front bumper.

[0082] Specifically, according to the imaging principle and the following formula, the distance of the camera relative to the front bumper is obtained, and the formula is as follows:

[0083] d=f*L / p;

[0084] In the formula, d is the distance of the camera relative to the front bumper, the unit is m, f is the focal length of the camera, the unit is pixel, L is the length of the slide plate, the unit is m, and p is the pixel distance of the two marker points on the extrinsic calibration image, the unit is pixel.

[0085] As can be seen from the above, based on the extrinsic calibration auxiliary support device, the camera extrinsic calibration can be performed before the vehicle drives, and manual calibration of the camera extrinsic calibration is not required during the driving of the vehicle, so that the camera extrinsic calibration is more convenient, real-time perception of the target position on the camera is realized, and the target can be visually perceived, thereby providing convenience for the development and testing of the automatic driving function.​​

[0086] Corresponding to the above-mentioned embodiments, the application also discloses a camera extrinsic parameter calibration device.

[0087] Referring to Figure 9 , the application discloses a structure diagram of a camera extrinsic parameter calibration device, which comprises:

[0088] The first acquisition unit 301 is configured to acquire an intrinsic matrix and distortion coefficients of a camera calibrated by using Zhang Zhengyou camera calibration method.

[0089] The intrinsic matrix comprises four parameters, which are pixel lengths of focal lengths in x-axis and y-axis directions of a vehicle coordinate system and pixel coordinates of image centers in x-axis and y-axis directions of the vehicle coordinate system.

[0090] In actual application, a monocular camera is prepared, the focal length of the monocular camera is adjusted and kept unchanged, and a fixed-focus camera can be preferably selected, and the image resolution output by the fixed-focus camera is relatively high. A black-and-white checkerboard is prepared, the checkerboard has consistent square grids, the number of inner corner points is as large as possible, and is generally more than 3 (high) x 5 (wide), and the board surface is flat; the calibration board in the calibration image is free from glare, reflected light or other obstacle shielding; 15-20 images of the checkerboard in each posture are shot, the shooting angle is preferably 30-75 degrees to ensure the image quality and increase the distinguishability, and the image is clear and free from ghosting; the Zhang Zhengyou camera calibration method is adopted, the image and the number of inner corner points of the checkerboard are input into a camera intrinsic calibration program, and the intrinsic matrix and the distortion coefficients of the camera are obtained.

[0091] The second acquisition unit 302 is configured to acquire a lateral distance amount and a distance-ground height amount of the camera relative to the vehicle collected by an extrinsic parameter calibration auxiliary support device.

[0092] The extrinsic parameter calibration unit 303 is configured to calibrate a camera rotation angle and a distance of the camera relative to a front bumper based on the intrinsic matrix, the distortion coefficients, the lateral distance amount and the distance-ground height amount, and complete calibration of the camera extrinsic parameters.

[0093] The camera extrinsic parameters comprise the camera rotation angle and the distance of the camera relative to the front bumper, the camera rotation angle comprises a yaw angle, a pitch angle and a roll angle, and the distance of the camera relative to the front bumper comprises three relative distances of a camera optical center relative to the vehicle.

[0094] In conclusion, the camera external parameter calibration device is disclosed, the intrinsic matrix and the distortion coefficient of the camera calibrated by Zhang Zhengyou camera calibration method are obtained, the lateral distance amount and the distance ground height amount of the camera relative to the vehicle collected by the external parameter calibration auxiliary support equipment are obtained, the camera rotation angle and the distance of the camera relative to the front bumper are calibrated based on the intrinsic matrix, the distortion coefficient, the lateral distance amount and the distance ground height amount, and the calibration of the camera external parameter is completed. The camera external parameter calibration device based on the external parameter calibration auxiliary support equipment can calibrate the camera external parameter before the vehicle drives, and manual calibration of the camera external parameter is not needed in the driving process of the vehicle, so that the camera external parameter calibration is more convenient, real-time sensing of the target position on the camera is realized, the target can be visually sensed, and the development and test of the automatic driving function are facilitated.

[0095] It should be noted that the specific working principles of the components in the device embodiment are described in the corresponding part of the method embodiment, which will not be repeated here.

[0096] The application further discloses a camera external parameter calibration system, which comprises: Figure 2 The external parameter calibration auxiliary support equipment in the illustrated embodiment and Figure 9 The camera external parameter calibration device in the illustrated embodiment. The process of the camera external parameter calibration system for calibrating the camera external parameter can be seen in the corresponding part of the above-mentioned embodiment, which will not be repeated here.

[0097] Finally, it should be noted that in this paper, relationship terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between the entities or operations. Moreover, the terms "include", "contain" or any other variants thereof are intended to cover non-exclusive inclusion, so that the process, method, article or equipment including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such process, method, article or equipment. Without more limitations, the element defined by the statement "including a" does not exclude the presence of other identical elements in the process, method, article or equipment including the element.

[0098] The various embodiments in the specification are described in a progressive manner, and each embodiment focuses on the difference from other embodiments, and the same or similar parts between various embodiments can be referred to each other.

[0099] The foregoing description of the disclosed embodiments enables a person skilled in the art to make or use the application. Modifications of these embodiments will occur to persons of skill in the art, and that the appended claims are intended to cover all such modifications that do not depart from the true spirit and scope of the application. Therefore, the application is not limited to the embodiments shown but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for calibrating extrinsic parameters of a camera, characterized in that, include: Obtain the intrinsic parameter matrix and distortion coefficients of the camera calibrated using Zhang Zhengyou's camera calibration method; The external parameter calibration auxiliary support device acquires the lateral distance and ground height of the camera relative to the vehicle, collected by the external parameter calibration auxiliary support device. The external parameter calibration auxiliary support device includes a base support and a sliding support, the sliding support being vertically fixedly mounted on the base support. A fixed-length sliding plate is provided on a crossbar of the sliding support, and a laser is installed at the center of the sliding plate. The crossbar is movable along the sliding support, and the sliding plate is movable along the crossbar. Acquiring the lateral distance and ground height of the camera relative to the vehicle, collected by the external parameter calibration auxiliary support device, includes: when the external parameter calibration auxiliary support... When the device is placed in a preset position such that the center of the laser beam emitted by the laser falls on the camera lens position, the height of the laser above the ground and the center offset of the laser relative to the crossbar are obtained; the height of the laser above the ground is determined as the distance from the ground, and the center offset is determined as the lateral distance; wherein, the preset position is: the base bracket is aligned with the front wheel of the vehicle, the surface formed by the sliding bracket is in contact with the front bumper of the vehicle, the base bracket is close to the two front wheels of the vehicle, the direction of the laser beam emitted by the laser is consistent with the direction of the foot of the base bracket, and the laser beam is at the same height as the camera lens; Based on the intrinsic parameter matrix, the distortion coefficients, the lateral distance, and the ground clearance, the camera rotation angle and the distance of the camera relative to the front bumper are calibrated to complete the calibration of the camera's extrinsic parameters, including: Acquire an extrinsic calibration image of the entire skateboard captured by the camera, wherein the skateboard is the mechanism for placing the laser in the extrinsic calibration auxiliary support device; Two marker points are marked at both ends of the skateboard image on the extrinsic calibration image; Based on the intrinsic parameter matrix, the distortion coefficient, the lateral distance, and the height above the ground, the value of the camera rotation angle is continuously changed, wherein the initial value of the camera rotation angle is 0, and the camera rotation angle includes: yaw angle, pitch angle, and roll angle; The two marker points on the extrinsic calibration image are dynamically moved to the half-height position of the image, and the two marker points are symmetrical about the longitudinal central axis of the extrinsic calibration image; Record the opposite value of the target camera rotation angle when the two marker points are located at half height of the image and the two marker points are symmetrical about the longitudinal central axis of the extrinsic parameter calibration image, and determine the opposite value as the camera rotation angle calibration result; Based on the pixel distance between the two marker points on the extrinsic calibration image and the length of the skateboard, the distance of the camera relative to the front bumper is obtained; including: Based on the imaging principle and the following formula, the distance between the camera and the front bumper is obtained, as follows: d = f × L / p; In the formula, d is the distance between the camera and the front bumper, f is the focal length of the camera, L is the length of the skateboard, and p is the pixel distance between the two marker points on the extrinsic calibration image.

2. The camera extrinsic parameter calibration method according to claim 1, characterized in that, The base support is also provided with a connector, which is used to extend the two legs of the base support.

3. The camera extrinsic parameter calibration method according to claim 1, characterized in that, A groove is provided in the middle of the skateboard, and the laser is installed in the groove.

4. The camera extrinsic parameter calibration method according to claim 1, characterized in that, The step of continuously changing the camera rotation angle based on the intrinsic parameter matrix, the distortion coefficient, the lateral distance, and the height above the ground includes: Based on the intrinsic parameter matrix, the distortion coefficient, the lateral distance, and the height above the ground, the value of the camera rotation angle is continuously changed by dragging the slider using a slider callback function.

5. The camera extrinsic parameter calibration method according to claim 1, characterized in that, Changing the yaw angle causes the marker point to move along the horizontal axis of the extrinsic calibration image; changing the pitch angle causes the marker point to move along the vertical axis of the extrinsic calibration image; changing the roll angle causes the two marker points to rotate around a preset center point of the extrinsic calibration image.

6. A camera extrinsic parameter calibration device, characterized in that, include: The first acquisition unit is used to acquire the intrinsic parameter matrix and distortion coefficients of the camera calibrated using the Zhang Zhengyou camera calibration method; The second acquisition unit is used to acquire the lateral distance and height above the ground of the camera relative to the vehicle, collected by the external parameter calibration auxiliary support device. The external parameter calibration auxiliary support device includes: a base support and a sliding support, wherein the sliding support is vertically fixedly installed on the base support; a fixed-length sliding plate is provided on the crossbar of the sliding support, and a laser is installed at the center of the sliding plate; wherein the crossbar can move along the sliding support, and the sliding plate can move along the crossbar; The extrinsic parameter calibration unit is used to calibrate the camera rotation angle and the distance of the camera relative to the front bumper based on the intrinsic parameter matrix, the distortion coefficient, the lateral distance, and the height above the ground, thereby completing the calibration of the camera's extrinsic parameters. The second acquisition unit is specifically used for: When the external parameter calibration auxiliary support device is placed in a preset position so that the center of the laser beam emitted by the laser falls on the camera lens position, the laser's height above the ground and the laser's center offset relative to the crossbar are obtained; the laser's height above the ground is determined as the distance from the ground, and the center offset is determined as the lateral distance; wherein, the preset position is: the base support is aligned with the front wheel of the vehicle, the surface formed by the sliding support is in contact with the front bumper of the vehicle, the base support is close to the two front wheels of the vehicle, the direction of the laser beam emitted by the laser is consistent with the foot direction of the base support, and the laser beam is at the same height as the camera lens; The external parameter calibration unit is specifically used for: Acquire an extrinsic calibration image of the entire skateboard captured by the camera, wherein the skateboard is the mechanism for placing the laser in the extrinsic calibration auxiliary support device; Two marker points are marked at both ends of the skateboard image on the extrinsic calibration image; Based on the intrinsic parameter matrix, the distortion coefficient, the lateral distance, and the height above the ground, the value of the camera rotation angle is continuously changed, wherein the initial value of the camera rotation angle is 0, and the camera rotation angle includes: yaw angle, pitch angle, and roll angle; The two marker points on the extrinsic calibration image are dynamically moved to the half-height position of the image, and the two marker points are symmetrical about the longitudinal central axis of the extrinsic calibration image; Record the opposite value of the target camera rotation angle when the two marker points are located at half height of the image and the two marker points are symmetrical about the longitudinal central axis of the extrinsic parameter calibration image, and determine the opposite value as the camera rotation angle calibration result; Based on the pixel distance between the two marker points on the extrinsic calibration image and the length of the skateboard, the distance of the camera relative to the front bumper is obtained; including: Based on the imaging principle and the following formula, the distance between the camera and the front bumper is obtained, as follows: d = f × L / p; In the formula, d is the distance between the camera and the front bumper, f is the focal length of the camera, L is the length of the skateboard, and p is the pixel distance between the two marker points on the extrinsic calibration image.

7. A camera extrinsic parameter calibration system, characterized in that, include: The external parameter calibration auxiliary support device according to claim 1 and the camera external parameter calibration device according to claim 6.

Images