Multi-sensor calibration system and method
By using the control terminal and the calibration board system in the calibration box, the joint calibration of the camera, IMU and LIDAR was achieved, which solved the problems of low efficiency and poor reliability in the existing technology and is suitable for mass production environment.
Patent Information
- Application Number
- CN202310154047.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2023-02-23
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2043-02-23
AI Technical Summary
Existing multi-sensor calibration techniques are inefficient and unreliable, and cannot achieve joint calibration of multiple sensors on the same system. In particular, the joint calibration process of cameras with IMUs and LiDARs needs to be performed separately, and the calibration results are not reliable enough.
A multi-sensor calibration system is provided, including a control terminal, a calibration box, and a fixed turntable. The control terminal controls the calibration plates on the fixed turntable by preset motion trajectories, so as to realize the joint calibration of cameras, cameras and IMUs, and cameras and LiDARs. Several calibration plates are set in the calibration box to form feature patterns, which supports the rapid calibration of multiple sensors.
It improves the efficiency and reliability of multi-sensor calibration, simplifies the operation process, is suitable for mass production solutions, and reduces calibration difficulty and cost.
Smart Images

Figure CN115908589B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of machine vision technology, and in particular to a multi-sensor calibration system and method. Background Art
[0002] With the advancement of technology, multi-sensor fusion solutions have become the mainstream direction of unmanned driving or simultaneous positioning and mapping technology, namely SLAM (Simultaneous Localization And Mapping) technology. Therefore, multi-sensor calibration has also become an indispensable and important link.
[0003] In the existing technology for multi-sensor calibration, only a single calibration plate can be used to calibrate a single or multiple cameras, or a single or multiple cameras can be calibrated at one time through multiple calibration plates. However, it needs to be performed statically and has high calibration requirements. The joint calibration process of the camera and the inertial measurement unit IMU (Inertial measurement unit) and the joint calibration process of the camera and the laser radar LIDAR (light detection and ranging) need to be performed separately. That is, a single calibration system is obviously unable to meet the existing multi-sensor calibration needs, and multiple systems need to be performed separately. Therefore, the calibration efficiency is low, and due to the differences in equipment between different systems, the reliability of the final calibration results is also poor. Summary of the invention
[0004] The present invention aims to provide a multi-sensor calibration system and method to solve the above-mentioned technical problems, realize the calibration of multiple sensors on the same system, effectively improve the efficiency and reliability of calibration, and be suitable for the calibration process of mass production solutions.
[0005] In order to solve the above technical problems, the present invention provides a multi-sensor calibration system, including a control terminal, a calibration box and a fixed turntable arranged in the calibration box and communicatively connected to the control terminal; a plurality of calibration plates are arranged in the calibration box; wherein:
[0006] The control terminal is used to preset the motion trajectory of the fixed turntable and transmit a control signal to the fixed turntable;
[0007] The fixed turntable is used to fix the camera;
[0008] Based on the control of the control terminal, the camera calibration, the camera and IMU joint calibration process, and the camera and LIDAR joint calibration process are realized on several calibration boards.
[0009] The above system only needs the control terminal to control the fixed turntable, and cooperates with several calibration plates set in the calibration box to realize camera calibration, camera and IMU joint calibration process and camera and LIDAR joint calibration process, and realizes the calibration of multiple sensors on the same system. It is simple to operate, effectively improves the efficiency and reliability of calibration, and is suitable for the calibration process of mass production solutions.
[0010] In the above scheme, the control terminal controls the movement of the camera through a preset motion trajectory. On the one hand, it can quickly complete the joint calibration process of the camera and IMU, and on the other hand, it can provide relatively sufficient stimulation to the IMU, making the calibration result more reliable.
[0011] Furthermore, the calibration box is a closed structure.
[0012] In the above scheme, since the calibration box is a closed structure, the camera fixed on the fixed turntable will fall inside the calibration box and be isolated from the outside. For a camera with a large field of view, it will not lose its field of view during shooting, and the calibration plate can be seen in the entire field of view, which greatly reduces the difficulty of calibration and improves the efficiency and operability of calibration.
[0013] Furthermore, the plurality of calibration plates each include a substrate and a characteristic pattern, and the characteristic pattern is a checkerboard, a QR code or a custom codec.
[0014] In the above scheme, the calibration plates of the calibration box can realize the rapid calibration of the camera, and can be realized without moving the camera under normal calibration conditions. At the same time, the characteristic patterns of the calibration plates can be set in various styles to improve the compatibility of the system.
[0015] Furthermore, the several calibration panels are all backlight calibration panels, and their control ends are all electrically connected to the control terminal; the control terminal is used to control the backlight conditions of the several backlight calibration panels to form characteristic patterns required for calibration on the corresponding backlight calibration panels.
[0016] In the above scheme, the calibration plate is set as a backlight calibration plate, which can directly form the characteristic pattern required for calibration on the backlight calibration plate through the control of the control terminal, avoiding the tedious process of manually replacing the calibration plate during the calibration process, and improving the efficiency of calibration. Secondly, the presence of backlight can make the camera's imaging quality higher, and can better reflect the lines, surfaces and reflectivity of the characteristic pattern, which is conducive to the joint calibration process of the camera and LIDAR.
[0017] The present invention provides a multi-sensor calibration method, which is implemented based on a multi-sensor calibration system and includes:
[0018] Determine calibration requirements;
[0019] Based on the calibration requirements, the fixed turntable is controlled according to the preset motion trajectory, and the camera fixed on the fixed turntable is calibrated on a plurality of preset calibration plates;
[0020] The calibration requirements include camera calibration, camera and IMU joint calibration process, and camera and LIDAR joint calibration process.
[0021] In the above scheme, only the control terminal is required to control the fixed turntable, and with the help of several calibration plates set in the calibration box, the camera calibration, the camera and IMU joint calibration process, and the camera and LIDAR joint calibration process can be realized, thus realizing the calibration of multiple sensors on the same system. The operation is simple, which effectively improves the efficiency and reliability of the calibration and is suitable for the calibration process of mass production schemes.
[0022] Furthermore, if the calibration requirement is for the calibration of a camera, the fixed turntable is controlled according to a preset motion trajectory, and the calibration of the camera fixed on the fixed turntable is realized on a plurality of preset calibration plates, specifically:
[0023] Control the fixed turntable to move according to a preset motion trajectory so that the camera can acquire the image of the calibration plate;
[0024] Determine the boundary line of the calibration plate based on the calibration plate image acquired by the camera, and segment different calibration plate images;
[0025] Perform encoding and decoding on the segmented calibration plate image to obtain the control points of each calibration plate;
[0026] The camera intrinsic parameters are calibrated based on the control point coordinates and the corresponding world coordinates.
[0027] Furthermore, if the calibration requirement is a joint calibration process of the camera and the IMU, the fixed turntable is controlled according to the preset motion trajectory, and the calibration of the camera fixed on the fixed turntable is realized on a preset number of calibration plates, specifically:
[0028] Control the movement of the fixed turntable according to the preset motion trajectory so that the camera can acquire the image of the calibration plate;
[0029] Determine the boundary line of the calibration plate based on the calibration plate image acquired by the camera, and segment different calibration plate images;
[0030] Encode and decode the segmented calibration plate image to obtain the control points of each calibration plate and calculate the camera extrinsic parameters;
[0031] Determine the camera pose according to the motion trajectory and initialize the nodes of the B-spline;
[0032] The observation model cost function is established based on the nodes of B-spline, and the external parameters of the camera and IMU are obtained after optimization to realize the joint calibration process of the camera and IMU.
[0033] Furthermore, if the calibration requirement is a joint calibration process of a camera and a LIDAR, the fixed turntable is controlled according to a preset motion trajectory, and the calibration of the camera fixed on the fixed turntable is realized on a preset number of calibration plates, specifically:
[0034] Control the fixed turntable to move according to a preset motion trajectory so that the camera can acquire the image of the calibration plate;
[0035] Determine the calibration plate plane and boundary line based on the calibration plate image acquired by the camera, and segment different calibration plate planes;
[0036] Based on the point cloud data obtained by the laser radar, the point cloud plane and the intersection line of the plane are extracted;
[0037] Based on the calibration plate plane, boundary line, point cloud plane and the intersection line of the plane, the joint optimization of camera and LIDAR point to line and point to surface is realized, and the joint calibration process of camera and LIDAR is realized.
[0038] The present invention provides an electronic device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned multi-sensor calibration method when executing the computer program.
[0039] The present invention provides a computer program product. When the computer program product is run on a terminal device, the terminal device implements the multi-sensor calibration method mentioned above. BRIEF DESCRIPTION OF THE DRAWINGS
[0040] Figure 1 A multi-sensor calibration system architecture diagram provided by one embodiment of the present invention;
[0041] Figure 2 A flow chart of a multi-sensor calibration method provided by one embodiment of the present invention;
[0042] Among them: 1. Control terminal; 2. Calibration box; 3. Fixed turntable. DETAILED DESCRIPTION
[0043] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0044] This embodiment provides a multi-sensor calibration system, including a control terminal 1, a calibration box 2, and a fixed turntable 3 disposed in the calibration box 2 and communicatively connected to the control terminal 1; the calibration box 2 is provided with a plurality of calibration plates; wherein:
[0045] The control terminal 1 is used to preset the motion trajectory of the fixed turntable 3 and transmit a control signal to the fixed turntable 3;
[0046] The fixed turntable 3 is used to fix the camera;
[0047] Based on the control of the control terminal 1, the calibration of the camera, the joint calibration process of the camera and IMU, and the joint calibration process of the camera and LIDAR are realized on several calibration boards.
[0048] This embodiment only requires the control terminal 1 to control the fixed turntable 3, and cooperates with several calibration plates set in the calibration box 2 to realize the camera calibration, the camera and IMU joint calibration process, and the camera and LIDAR joint calibration process, thereby realizing the calibration of multiple sensors on the same system. The operation is simple, and the efficiency and reliability of the calibration are effectively improved, which is suitable for the calibration process of the mass production solution.
[0049] In this embodiment, the control terminal 1 controls the movement of the camera through a preset motion trajectory, which can quickly complete the joint calibration process of the camera and IMU on the one hand, and fully stimulate the IMU on the other hand, making the calibration result more reliable. For example, during the camera IMU calibration, the IMU is fully stimulated due to the presence of the fixed turntable 3, the camera will not lose FOV, and the posture of each frame can be solved.
[0050] It should be noted that the above system can also perform single or combined calibration of other sensors according to needs. It integrates multi-sensor calibration into one, can flexibly and efficiently realize multi-sensor calibration, and can efficiently complete the calibration of any camera's internal and external parameters. Moreover, each inner surface of the calibration box 2 is provided with a corresponding calibration plate to ensure that the calibration box 2 has a 270° calibration space, which can improve the reliability of large FOV camera calibration.
[0051] It should be noted that the fixed turntable 3 can be implemented by a three-degree-of-freedom three-axis turntable, and the control terminal 1 can be implemented by a PC.
[0052] Furthermore, the calibration box 2 is a closed structure.
[0053] In this embodiment, since the calibration box 2 is a closed structure, the camera fixed on the fixed turntable 3 will fall inside the calibration box 2 and be isolated from the outside. For a camera with a large field of view, it will not lose its field of view during the shooting process. For example, after a camera with a field of view exceeding 120° is fixed in the calibration box 2, the calibration plate can be seen in the entire field of view, which greatly reduces the difficulty of calibration and improves the efficiency and operability of calibration.
[0054] Furthermore, the plurality of calibration plates each include a substrate and a characteristic pattern, and the characteristic pattern is a checkerboard, a QR code or a custom codec.
[0055] In this embodiment, the calibration plates of the calibration box 2 can realize the rapid calibration of the camera, which can be realized without moving the camera in general calibration situations. For example, after fixing the camera in the calibration box 2, the camera calibration can be performed by taking a photo of the inside of the calibration box 2. At the same time, the characteristic patterns of the calibration plates can be set in various styles to improve the compatibility of the system.
[0056] Furthermore, the several calibration panels are all backlight calibration panels, and their control ends are all electrically connected to the control terminal 1; the control terminal 1 is used to control the backlight conditions of the several backlight calibration panels to form characteristic patterns required for calibration on the corresponding backlight calibration panels.
[0057] In this embodiment, the calibration plate is set as a backlight calibration plate, which can directly form the characteristic pattern required for calibration on the backlight calibration plate through the control of the control terminal 1, avoiding the tedious process of manually replacing the calibration plate during the calibration process, and improving the efficiency of calibration. Secondly, the presence of backlight can make the imaging quality of the camera higher, and because the interior of the calibration box 2 is composed of a surface structure, its characteristic pattern is generally composed of different colors alternating, which can better reflect the line, surface and reflectivity of the characteristic pattern, which is conducive to the joint calibration process of the camera and LIDAR. For example, when calibrating the camera and the laser radar, different surface constraints and line constraints can be provided due to the obvious structure of the calibration box.
[0058] In order to better reflect the technical points of the present invention, this embodiment compares and analyzes the technical points of the present invention with the prior art to highlight the technical advantages of the present invention. Specifically, it includes:
[0059] In the prior art, when calibrating a camera using a single calibration plate, at least three pictures need to be taken in three different postures. The multi-sensor calibration system provided in this embodiment can obtain multiple calibration plate pictures in one shooting, which greatly improves the calibration efficiency and simplifies the calibration steps.
[0060] In the prior art, when a single calibration board is used to perform joint calibration of the camera and IMU, the movement space of the camera and IMU will be limited, and they can only move in a space where the camera can see the entire calibration board, which can easily lead to insufficient system excitation. In the calibration box 2 provided by the present invention, the camera can see the calibration board when it moves arbitrarily, which makes the excitation of the system more effective and the calibration result more reliable.
[0061] In the prior art, when a single calibration plate is used for joint camera and LIDAR calibration, since only one effective constraint of one surface can be provided each time and the provided edge line feature constraint is inaccurate, multiple shots are required to obtain data to complete the calibration. The calibration box 2 provides line feature constraints of three surfaces and three surface intersection lines each time. Such constraints are more efficient and reliable, greatly improving the efficiency of calibration and simplifying the calibration steps.
[0062] See also Figure 2 This embodiment provides a multi-sensor calibration method, which is implemented based on a multi-sensor calibration system and includes:
[0063] S1: Determine calibration requirements;
[0064] S2: Based on the calibration requirements, the fixed turntable is controlled according to the preset motion trajectory, and the camera fixed on the fixed turntable is calibrated on a plurality of preset calibration plates;
[0065] The calibration requirements include camera calibration, camera and IMU joint calibration process, and camera and LIDAR joint calibration process.
[0066] In this embodiment, only the control terminal is required to control the fixed turntable, and with the help of several calibration plates provided in the calibration box, the camera calibration, the camera and IMU joint calibration process, and the camera and LIDAR joint calibration process can be realized, thereby realizing the calibration of multiple sensors on the same system. The operation is simple, and the efficiency and reliability of the calibration are effectively improved, which is suitable for the calibration process of mass production solutions.
[0067] Furthermore, if the calibration requirement is for the calibration of a camera, the fixed turntable is controlled according to a preset motion trajectory, and the calibration of the camera fixed on the fixed turntable is realized on a plurality of preset calibration plates, specifically:
[0068] Control the movement of the fixed turntable according to the preset motion trajectory so that the camera can acquire the image of the calibration plate;
[0069] Determine the boundary line of the calibration plate based on the calibration plate image acquired by the camera, and segment different calibration plate images;
[0070] Perform encoding and decoding on the segmented calibration plate image to obtain the control points of each calibration plate;
[0071] The camera intrinsic parameters are calibrated based on the control point coordinates and the corresponding world coordinates.
[0072] It should be noted that before calibrating the camera's intrinsic parameters, it is necessary to calibrate the calibration plate's extrinsic parameters, that is, to restore the extrinsic parameters between the calibration plates through a large number of captured images, and then calibrate them as calibration configuration parameters.
[0073] Furthermore, the calibration plate image acquired based on the camera determines the boundary line of the calibration plate and segments different calibration plate images, specifically:
[0074] The calibration plate image acquired by the camera is locally binarized (using a 20x20 window to calculate, counting the inter-class variance of pixel values, distinguishing between background and foreground, and binarizing according to the foreground and background thresholds) to ensure the local characteristics of the calibration plate image for easy image segmentation and decoding of the calibration plate; straight line detection is performed on the binary image to obtain the boundary lines of the calibration plate, and each calibration plate captured is separated according to the intersection of the boundary lines, and different calibration plate images are segmented.
[0075] Furthermore, based on the control point coordinates and the corresponding world coordinates, the camera intrinsic parameters are calibrated, specifically:
[0076] The camera intrinsic parameters are initialized and optimized using the control point coordinates and their corresponding world coordinates. This process mainly treats the split single calibration plate as an independent observation. The four calibration plates are four independent observations. The first camera calibration is performed using the four independent observations, and the first calibration is defined as local calibration. According to the first camera calibration result, the world coordinate system is established using the 0th calibration plate. The world coordinates of all calibration plates are transferred to the 0th calibration plate using the external parameters between the calibration plates. The second calibration is performed under constraints to obtain the external parameters of each image taken and the intrinsic parameters of the camera. The second calibration is defined as global calibration. The perspective transformation and the external parameters of each picture taken are used to perform an inverse perspective transformation to change the image into a normal view. The calibration process is repeated multiple times to finally complete the calibration of the camera intrinsic parameters.
[0077] Furthermore, if the calibration requirement is a joint calibration process of the camera and the IMU, the fixed turntable is controlled according to the preset motion trajectory, and the calibration of the camera fixed on the fixed turntable is realized on a preset number of calibration plates, specifically:
[0078] Control the fixed turntable to move according to a preset motion trajectory so that the camera can acquire the image of the calibration plate;
[0079] Determine the boundary line of the calibration plate based on the calibration plate image acquired by the camera, and segment different calibration plate images;
[0080] Encode and decode the segmented calibration plate image to obtain the control points of each calibration plate and calculate the camera extrinsic parameters;
[0081] Determine the camera pose according to the motion trajectory and initialize the nodes of the B-spline;
[0082] The observation model cost function is established based on the nodes of B-spline, and the external parameters of the camera and IMU are obtained after optimization to realize the joint calibration process of the camera and IMU.
[0083] Furthermore, if the calibration requirement is a joint calibration process of a camera and a LIDAR, the fixed turntable is controlled according to a preset motion trajectory, and the calibration of the camera fixed on the fixed turntable is realized on a preset number of calibration plates, specifically:
[0084] Control the fixed turntable to move according to a preset motion trajectory so that the camera can acquire the image of the calibration plate;
[0085] Determine the calibration plate plane and boundary line based on the calibration plate image acquired by the camera, and segment different calibration plate planes;
[0086] Based on the point cloud data obtained by the laser radar, the point cloud plane and the intersection line of the plane are extracted;
[0087] Based on the calibration plate plane, boundary line, point cloud plane and the intersection line of the plane, the joint optimization of camera and LIDAR point to line and point to surface is realized, and the joint calibration process of camera and LIDAR is realized.
[0088] The above-mentioned embodiment satisfies the condition that multiple sensor calibration tasks can be completed in one system in an actual production line, has greater applicability to different sensors, and greatly saves the space and cost required for calibration.
[0089] This embodiment provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned multi-sensor calibration method when executing the computer program.
[0090] This embodiment provides a computer program product. When the computer program product is run on a terminal device, the terminal device implements the multi-sensor calibration method described above.
[0091] The above is a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements and modifications without departing from the principle of the present invention. These improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A multi-sensor calibration system, comprising a control terminal, a calibration box and a fixed turntable arranged in the calibration box and communicatively connected to the control terminal; characterized in that: The inner surface of the calibration box is provided with corresponding calibration plates; the calibration box is a closed structure; wherein: The control terminal is used to control the motion trajectory of the preset fixed turntable and transmit a control signal to the fixed turntable; The fixed turntable is used to fix the camera; Based on the control of the control terminal, the calibration of the camera, the joint calibration of the camera and IMU, and the joint calibration of the camera and LIDAR are realized on the calibration board; The camera calibration is specifically as follows: controlling the fixed turntable to move according to a preset motion trajectory so that the camera acquires the calibration plate image; determining the boundary line of the calibration plate based on the calibration plate image acquired by the camera, and segmenting different calibration plate images; encoding and decoding the calibration plate images obtained after segmentation to obtain the control points of each calibration plate; calibrating the camera internal parameters based on the control point coordinates and the corresponding world coordinates; The joint calibration process of the camera and IMU is specifically as follows: controlling the fixed turntable to move according to a preset motion trajectory so that the camera can acquire the calibration plate image; determining the boundary line of the calibration plate based on the calibration plate image acquired by the camera, and segmenting different calibration plate images; encoding and decoding the calibration plate image obtained after segmentation to obtain the control point of each calibration plate and calculate the camera extrinsic parameters; determining the camera pose according to the motion trajectory and initializing the nodes of the B-spline; establishing the observation model cost function based on the nodes of the B-spline, optimizing and obtaining the extrinsic parameters of the camera and the IMU, and realizing the joint calibration process of the camera and the IMU; The joint calibration process of the camera and LIDAR is specifically as follows: controlling the fixed turntable to move according to a preset motion trajectory so that the camera can acquire the calibration plate image; determining the calibration plate plane and boundary line based on the calibration plate image acquired by the camera, and segmenting different calibration plate planes; extracting the point cloud plane and the intersection line of the plane based on the point cloud data acquired by the laser radar; realizing the joint optimization of the camera and LIDAR from point to line and from point to surface based on the calibration plate plane, boundary line, point cloud plane and the intersection line of the plane, and realizing the joint calibration process of the camera and LIDAR; The calibration plates are all backlight calibration plates, and their control ends are all electrically connected to the control terminal; the control terminal is used to control the backlight conditions of the backlight calibration plates to form characteristic patterns required for calibration on the corresponding backlight calibration plates.
2. A multi-sensor calibration system according to claim 1, characterized in that: The calibration plates all include a substrate and a characteristic pattern, and the characteristic pattern is a checkerboard, a two-dimensional code or a custom codec.
3. A multi-sensor calibration method, characterized in that: A multi-sensor calibration system according to claim 1 or 2 is implemented, comprising: Determine calibration requirements; The calibration requirements include camera calibration, camera and IMU joint calibration process, and camera and LIDAR joint calibration process; If the calibration requirement is for camera calibration, the fixed turntable is controlled according to the preset motion trajectory, and the camera fixed on the fixed turntable is calibrated on the preset calibration board, specifically: Control the movement of the fixed turntable according to the preset motion trajectory so that the camera can acquire the image of the calibration plate; Determine the boundary line of the calibration plate based on the calibration plate image acquired by the camera, and segment different calibration plate images; The calibration plate image obtained after segmentation is encoded and decoded to obtain the control points of each calibration plate; Calibrate the camera intrinsic parameters based on the control point coordinates and the corresponding world coordinates; If the calibration requirement is a joint calibration process of the camera and the IMU, the fixed turntable is controlled according to the preset motion trajectory, and the calibration of the camera fixed on the fixed turntable is realized on the preset calibration board, specifically: Control the movement of the fixed turntable according to the preset motion trajectory so that the camera can acquire the image of the calibration plate; Determine the boundary line of the calibration plate based on the calibration plate image acquired by the camera, and segment different calibration plate images; The calibration plate image obtained after segmentation is encoded and decoded to obtain the control points of each calibration plate and calculate the camera extrinsic parameters; Determine the camera pose according to the motion trajectory and initialize the nodes of the B-spline; The observation model cost function is established based on the nodes of B-spline, and the external parameters of the camera and IMU are obtained after optimization to realize the joint calibration process of the camera and IMU; If the calibration requirement is a joint calibration process of the camera and LIDAR, the fixed turntable is controlled according to the preset motion trajectory, and the camera fixed on the fixed turntable is calibrated on the preset calibration board, specifically: Control the movement of the fixed turntable according to the preset motion trajectory so that the camera can acquire the image of the calibration plate; Determine the calibration plate plane and boundary line based on the calibration plate image acquired by the camera, and segment different calibration plate planes; Based on the point cloud data obtained by the laser radar, the point cloud plane and the intersection line of the plane are extracted; Based on the calibration plate plane, boundary line, point cloud plane and the intersection line of the plane, the joint optimization of camera and LIDAR point to line and point to surface is realized, and the joint calibration process of camera and LIDAR is realized.
4. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the multi-sensor calibration method as claimed in claim 3 is implemented.
5. A computer program product, characterized in that When the computer program product runs on a terminal device, the terminal device executes a multi-sensor calibration method as claimed in claim 3.
