Method and device for determining error information of calibration external parameter value of camera

Abstract

The embodiment of the invention provides a camera calibration external parameter value error information determination method and device, and the method comprises the steps: obtaining a true value external parameter value and a lane line image of a camera, and enabling the lane line image to be used for determining the ground plane coordinates of a lane line group; adjusting a calibration external parameter value of the camera based on the true value external parameter value to obtain an adjustment result; performing imaging operation on the lane line group according to the truth value external parameter value and the adjustment result to generate a projection image of the lane line group, the projection image being used for determining ground plane coordinates of the imaged lane line group; and comparing the projection image with the lane line image to obtain error information corresponding to the calibration external parameter value. According to the method and the device, the technical problem that the calibration external parameter value error information is difficult to reversely deduce from the lane line group index information due to the fact that the specific influence of the camera calibration external parameter value error information on the lane line group index information in the lane line aerial view cannot be quantitatively analyzed in the related technology is solved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of intelligent driving technology, and in particular to a method and device for determining error information of extrinsic parameter values of a camera calibration. Background Art

[0002] In intelligent driving technology, high-precision camera calibration is crucial for ensuring stable system operation and improving driving safety. Calibration of camera extrinsic parameters aims to accurately determine the camera's spatial position and posture relative to the vehicle and other environmental objects. This process is crucial for key functions such as augmented reality mapping, path planning, and obstacle detection.

[0003] Currently, error evaluation for camera extrinsic calibration primarily relies on reprojection error evaluation and lane line bird's-eye view projection. However, the reprojection error evaluation method cannot effectively distinguish which specific extrinsic parameter value causes a large error in the calibration extrinsic parameter value, and the lane line bird's-eye view projection method cannot quantify the impact of the calibration extrinsic parameter error on various metrics related to lane line endpoints. As a result, it is impossible to infer the error values of the individual parameters in the calibration extrinsic parameter value from these metrics related to lane line endpoints.

[0004] To address the above-mentioned problems, no effective solutions have been proposed so far. Summary of the Invention

[0005] An embodiment of the present application provides a method for determining camera calibration extrinsic parameter value error information, aiming to solve the technical problem in related technologies that it is difficult to reversely infer calibration extrinsic parameter value error information from lane line group indicator information due to the inability to quantitatively analyze the specific impact of camera calibration extrinsic parameter value error information on lane line group indicator information in a lane line bird's-eye view map.

[0006] According to one embodiment of the present application, a method for determining error information of camera calibration extrinsic parameter values is provided, including: obtaining true extrinsic parameter values and a lane line image of a camera, wherein the lane line image is used to determine the ground plane coordinates of a lane line group; adjusting the calibration extrinsic parameter values of the camera based on the true extrinsic parameter values to obtain an adjustment result; performing an imaging operation on the lane line group according to the true extrinsic parameter values and the adjustment result to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging; and comparing the projection image with the lane line image to obtain error information corresponding to the calibration extrinsic parameter values.

[0007] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: by imaging the lane line endpoints in the camera system according to the true extrinsic parameter values of the camera, and then using the adjusted calibration extrinsic parameter values to project the lane line endpoints in the image onto the ground plane, a projection image corresponding to the lane line image is generated, thereby achieving independent quantification of the influence of each parameter error of the camera calibration extrinsic parameter value, not only intuitively demonstrating the influence of the calibration extrinsic parameter value error on the lane line presentation effect, but also achieving independent quantitative analysis of each specific parameter error, thereby improving the accuracy and efficiency of the camera extrinsic parameter value calibration, and further solving the technical problem in the related art that it is difficult to reversely infer the calibration extrinsic parameter value error information from the lane line group indicator information due to the inability to quantitatively analyze the specific influence of the camera calibration extrinsic parameter value error information on the lane line group indicator information in the lane line bird's-eye view.

[0008] Optionally, obtaining a lane line image includes: obtaining lane line parameters and lane parameters, wherein the lane line parameters include the width, length, number and starting end position information of the lane line, and the lane parameters include the width, length, number and position information of the lane; generating a lane line image based on the lane line parameters and lane parameters.

[0009] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: a lane line image that complies with national standards, has high complexity and practicality, and contains rich lane line information. It not only conforms to the common layout of actual road traffic, enhances the practicality and authenticity of the solution, but also increases the complexity of the lane line group through its specific layout and size, providing more data points for subsequent analysis, thereby helping to improve the accuracy of determining the error of the calibration external parameter value.

[0010] Optionally, adjusting the calibrated extrinsic parameter value of the camera based on the true extrinsic parameter value to obtain the adjustment result includes: obtaining an accuracy threshold of the calibrated extrinsic parameter value; adjusting at least one deflection angle parameter in the true extrinsic parameter value according to the accuracy threshold to obtain the deflection angle parameter in the adjustment result.

[0011] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: not only the accuracy range of the deflection angle is taken into account, but also a hierarchical analysis is performed on the impact of the deviation, from a single parameter to a composite parameter, and then to the simultaneous action of three parameters. It can effectively separate the influence of different parameters, clarify the specific effect of each parameter on the lane line group index information, and after clarifying the degree of coupling between the parameters, it can adjust the parameters more specifically during calibration, avoiding invalid or inefficient adjustments, thereby improving the overall efficiency and final accuracy of the calibration.

[0012] Optionally, the translation value in the adjustment result is set to the translation value of the true extrinsic parameter value.

[0013] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: simplifying the camera calibration process, reducing the number of parameters that need to be adjusted, thereby avoiding additional adjustment or optimization of the translation amount during the calibration process, thereby reducing the complexity and uncertainty of the calibration process, and making the calibration process more efficient and controllable while ensuring the calibration accuracy.

[0014] Optionally, imaging operations are performed on the lane line group based on the true extrinsic parameter values and the adjustment results to generate a projection image of the lane line group, including: simulating imaging of the lane line group based on the true extrinsic parameter values to obtain imaging results; and projecting the imaging results onto a ground plane coordinate system using the adjustment results to generate a projection image of the lane line group.

[0015] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: by separating the effects of true value extrinsic parameters and calibrated extrinsic parameters, high precision can be maintained in the simulation imaging process, and the direct impact of the calibration extrinsic parameter deviation on the lane line projection result can be clearly demonstrated in the projection stage, which helps to accurately identify and isolate the source of the error, and the projection image generated by projecting to the ground plane coordinate system intuitively shows the changes in the shape and position of the lane line group caused by the deviation of the calibration extrinsic parameter value, providing an intuitive basis for subsequent error analysis and correction.

[0016] Optionally, the projection image is compared with the lane line image to obtain error information corresponding to the calibrated external parameter value, including: visualizing the projection image and the lane line image according to the horizontal pixel ratio and the vertical pixel ratio to obtain the position information of the lane line group in the projection image and the lane line group in the lane line image in the same bird's-eye view; and determining the error information corresponding to the calibrated external parameter value based on the position information.

[0017] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: the defined horizontal and vertical pixel ratios ensure that the visualization in the bird's-eye view can accurately reflect the actual size, thereby improving the accuracy of error information determination, and by visually comparing the lane line groups in the bird's-eye view, the position differences of the lane line groups are intuitively displayed, and linking the visualization result display with the specific extrinsic calibration parameter value can quickly locate the source of the error and understand the actual impact of the parameter deviation on the projection effect of the lane line group.

[0018] Optionally, determining error information corresponding to the extrinsic parameter value based on the position information includes:

[0019] The lane line group indicator information is determined based on the position information, wherein the lane line group indicator information includes the lane line length and lane width in the lane line image in the bird's-eye view, the distance information between the lane line and the camera in the projection image, and the yaw angle information of the lane line; the error information of the calibrated external parameter value is determined based on the lane line group indicator information.

[0020] The above-mentioned optional embodiments of the present application can achieve the following beneficial effects: by converting position information into intuitive lane line group index information, it not only achieves accurate quantification of the error of the external parameter value of the camera calibration, enhances the interpretability of data analysis and the operator's understanding of the results, but also, by analyzing the specific impact of the error on the lane line group index and the interaction between parameters, thereby improving the accuracy and efficiency of error identification, providing quantitative data and analysis results for the optimization of camera calibration in the intelligent driving system, thereby helping to improve the reliability of the system.

[0021] According to one embodiment of the present application, a device for determining the error of the calibration extrinsic parameter value of a camera is also provided, including: an acquisition module for acquiring the true extrinsic parameter value and the lane line image of the camera, wherein the lane line image is used to determine the ground plane coordinates of the lane line group; an adjustment module for adjusting the calibration extrinsic parameter value of the camera based on the true extrinsic parameter value to obtain an adjustment result; a generation module for performing an imaging operation on the lane line group according to the true extrinsic parameter value and the adjustment result to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging; a determination module for comparing the projection image with the lane line image to obtain error information corresponding to the calibration extrinsic parameter value.

[0022] Optionally, the acquisition module is also used to: obtain lane line parameters and lane parameters, wherein the lane line parameters include the width, length, number and starting end position information of the lane line, and the lane parameters include the width, length, number and position information of the lane; generate a lane line image based on the lane line parameters and lane parameters.

[0023] Optionally, the adjustment module is further used to: obtain a precision threshold of the calibration extrinsic parameter value; and adjust at least one deflection angle parameter in the true value extrinsic parameter value according to the precision threshold to obtain the deflection angle parameter in the adjustment result.

[0024] Optionally, the translation value in the adjustment result is set to the translation value of the true extrinsic parameter value.

[0025] Optionally, the generation module is further used to: simulate imaging of the lane line group based on the true external parameter value to obtain an imaging result; use the adjustment result to project the imaging result to the ground plane coordinate system to generate a projection image of the lane line group.

[0026] Optionally, the determination module is also used to: visualize the projection image and the lane line image according to the horizontal pixel ratio and the vertical pixel ratio to obtain the position information of the lane line group in the projection image and the lane line group in the lane line image in the same bird's-eye view; and determine the error information corresponding to the calibrated external parameter value based on the position information.

[0027] Optionally, the determination module is also used to: determine the lane line group index information based on the position information, wherein the lane line group index information includes the lane line length and lane width in the lane line image in the bird's-eye view, the distance information between the lane line and the camera in the projection image, and the yaw angle information of the lane line; determine the error information of the calibrated external parameter value based on the lane line group index information.

[0028] According to another aspect of an embodiment of the present application, an electronic device is provided, comprising a processor; and a memory for storing processor-executable instructions; wherein the processor is configured to execute instructions to implement the above-mentioned method for determining the error information of the external parameter value of the camera calibration.

[0029] According to another aspect of an embodiment of the present application, a computer-readable storage medium is provided, in which a computer program is stored, wherein the computer program is configured to execute the above-mentioned method for determining the error information of the external parameter value of the camera calibration when the computer program is executed by a processor.

[0030] According to another aspect of an embodiment of the present application, a computer program product is provided, including a computer program, which implements the above-mentioned method for determining the error information of the camera calibration external parameter value when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flow chart of a method for determining error information of camera calibration extrinsic parameter values provided by one embodiment of the present application;

[0032] Figure 2 1 is a schematic diagram of a method for determining error information of extrinsic parameter values of a camera calibration provided by an embodiment of the present application;

[0033] Figure 3A is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0034] Figure 3B is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0035] Figure 3C is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0036] Figure 4A is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0037] Figure 4B is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0038] Figure 4C is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0039] Figure 5 is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0040] Figure 6 is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0041] Figure 7 is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application;

[0042] Figure 8 This is a structural diagram of a device for determining error information of camera calibration extrinsic parameter values provided in an embodiment of the present application;

[0043] Figure 9 This is a structural diagram of an electronic device provided in one embodiment of the present application. DETAILED DESCRIPTION

[0044] In order to make the technical problems, technical solutions and beneficial effects solved by this application more clearly understood, this application is further described in detail below in conjunction with the embodiments. It should be understood that the specific embodiments described herein are only used to explain this application and are not intended to limit this application.

[0045] In related technologies, during the offline calibration of camera extrinsic parameter values, there are two main ways to evaluate the error of the calibration extrinsic parameter values:

[0046] (1) Reprojection error evaluation: This method provides an indicator to quantify the error. Specifically, in the calibration task, we first use the known 3D point coordinates and their corresponding camera parameters (intrinsic parameter values and calibrated extrinsic parameter values) to project the 3D point onto the image plane. Then, we compare the pixel positions of the projected point with the feature points in the actual image and calculate the difference between the two, which is the reprojection error. Since the core goal of extrinsic parameter calibration is to minimize the reprojection error, using it to measure the calibration quality of the extrinsic parameter values can directly reflect the quality of the calibration results. However, a significant drawback of this method is that the reprojection error, as a comprehensive result, cannot reveal the impact of each of the six independent parameters (i.e., three deflection angles and three translations) in the calibration extrinsic parameter values on the overall error. This means that although the reprojection error can tell us the size of the overall error, it cannot guide us to accurately adjust the specific extrinsic parameter values to further optimize the extrinsic parameter calibration results.

[0047] (2) Lane line bird's-eye view projection evaluation: Compared with the quantitative analysis of reprojection error, lane line bird's-eye view projection provides qualitative error feedback. By combining the camera's intrinsic parameter values and the calibration extrinsic parameters, the lane image captured by the camera is converted into a perspective-corrected bird's-eye view. The relative relationship between lane lines, such as parallelism, spacing, and overall layout, can be visually checked. This inspection method helps to evaluate the overall effect of the extrinsic parameter value calibration, especially for understanding the camera's field of view coverage and position deviation. However, relying solely on visual inspection of the bird's-eye view to analyze the specific errors of the extrinsic parameter values, and based on the quantitative indicators of the lane line bird's-eye view endpoints, it is used to accurately evaluate the error of each of the six independent extrinsic parameter values. This means that although we can see from the bird's-eye view whether the shape and position of the lane line are as expected, it is not enough to fine-tune each extrinsic parameter value to achieve the best calibration effect by visual observation alone. There is a lack of means to quantify the specific parameter errors.

[0048] An embodiment of the present application provides a method for determining error information of camera calibration extrinsic parameter values, including: obtaining true extrinsic parameter values and a lane line image of a camera, wherein the lane line image is used to determine the ground plane coordinates of a lane line group; adjusting the calibration extrinsic parameter values of the camera based on the true extrinsic parameter values to obtain an adjustment result; performing an imaging operation on the lane line group based on the true extrinsic parameter values and the adjustment result to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging; and comparing the projection image with the lane line image to obtain error information corresponding to the calibration extrinsic parameter values.

[0049] The above-mentioned method for determining the camera calibration extrinsic parameter value error information provided in the embodiment of the present application achieves the following technical effects: by imaging the lane line endpoints in the camera system according to the true extrinsic parameter values of the camera, and then using the adjusted calibration extrinsic parameter values to project the lane line endpoints in the image onto the ground plane, a projection image corresponding to the lane line image is generated, thereby achieving independent quantification of the influence of each parameter error of the camera calibration extrinsic parameter value, not only intuitively demonstrating the influence of the calibration extrinsic parameter value error on the lane line presentation effect, but also achieving independent quantitative analysis of each specific parameter error, thereby improving the accuracy and efficiency of the camera extrinsic parameter value calibration, and further solving the technical problem in the related art that it is difficult to reversely infer the calibration extrinsic parameter value error information from the lane line group indicator information due to the inability to quantitatively analyze the specific influence of the camera calibration extrinsic parameter value error information on the lane line group indicator information in the lane line bird's-eye view map.

[0050] Example 1

[0051] This application embodiment provides a method for determining error information of camera calibration external parameter values, please refer to Figure 1 , Figure 1 This is a flowchart of a method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, the method comprising the following steps:

[0052] Step S10, obtaining the true extrinsic parameter values and lane line images of the camera, wherein the lane line images are used to determine the ground plane coordinates of the lane line group;

[0053] In step S10, the true extrinsic parameter value is used to characterize the actual installation extrinsic parameter of the camera, which is determined based on the relevant extrinsic parameter value of the camera.

[0054] Specifically, the errors that may occur during the actual installation of the camera are taken into account, especially the installation tolerances of the three deflection angles (Pitch angle, Roll angle, and Yaw angle). According to the design specifications of the camera's external parameters, the tolerances of the three deflection angles are set to ±3 degrees, that is, the deflection angle of the camera may fluctuate within the range of ±3 degrees during actual installation. Therefore, any value within this tolerance range will be selected as the deflection angle in the true external parameter value to simulate the actual installation posture of the camera. For the three installation translations of the camera, considering the accuracy of the manufacturing and installation process, the deviation between the translation and the designed installation position is small. Therefore, the translation in the true external parameter value is designed to be the translation of the designed installation position. This assumption is based on the controllability of translation errors in actual scenarios, and also avoids unnecessary complexity in the evaluation process.

[0055] Furthermore, when constructing the true extrinsic parameters, two camera placement strategies can be set: placing the camera directly above the origin of the geodetic coordinate system. This layout simplifies subsequent coordinate transformations and calculations, making it easier to intuitively understand and operate. Alternatively, the camera can be offset horizontally while maintaining its longitudinal coordinates aligned with the origin of the geodetic coordinate system. This layout is closer to actual application scenarios, especially when the camera is mounted off-center on the vehicle, as it more realistically reflects its spatial position and perspective.

[0056] For example, the true extrinsic parameter values of the camera are shown in Table 1. As shown in Table 1, the true extrinsic parameter values are set to: Pitch is -90°, Roll is 0°, Yaw is -90°, and the camera height is set to 1.5m, where the installation tolerances of all settings are assumed to be zero.

[0057] Table 1: Setting of out-of-truth parameter values

[0058] pitch(deg) -90 roll(deg) 0 yaw(deg) -90 Camera height (m) 1.5

[0059] The lane image is used to represent the standard pattern for the design of virtual and real lane lines, and to simulate the resulting lane group. The lane group's feature points, namely the four corner points of each lane segment, are extracted from the lane image, and their coordinates in the ground plane coordinate system are determined.

[0060] Step S12, adjusting the calibrated extrinsic parameter values of the camera based on the true extrinsic parameter values to obtain an adjustment result;

[0061] Specifically, based on the camera's true extrinsic parameter values—that is, the idealized values for the camera's extrinsic parameters (including but not limited to rotation angle and position offset)—a series of adjustments are made to the existing calibration extrinsic parameter values. This is done to simulate various deviations that may occur during camera calibration in real-world applications, resulting in a series of adjustment results.

[0062] Step S14: performing an imaging operation on the lane line group based on the true extrinsic parameter values and the adjustment results to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging;

[0063] Specifically, based on the true extrinsic parameter values and the adjustment results, the lane line group is simulated and imaged to generate a projected image. This process involves converting the lane line group points from the earth coordinate system to the camera's true coordinate system, performing the necessary normalization, projecting the points into the image coordinate system using the intrinsic parameter matrix, and converting the image coordinates back to the calibration coordinate system through an inverse transformation. Finally, the adjusted calibration extrinsic parameters are used again to project the calibration coordinates back into the earth coordinate system, taking special care to ensure that the projected points fall exactly on the ground plane (i.e., when z = 0).

[0064] Step S16: Compare the projection image with the lane line image to obtain error information corresponding to the calibrated extrinsic parameter value.

[0065] Specifically, based on the projection image, the projection image is compared with the lane line image, and the error information of the calibration extrinsic parameter value is quantified through image comparison and analysis.

[0066] Based on the above steps S10 to S16, the camera's true extrinsic parameter values and lane line images are obtained, and the camera's calibrated extrinsic parameter values are adjusted based on the true extrinsic parameter values to obtain an adjustment result. Furthermore, an imaging operation is performed on the lane line group based on the true extrinsic parameter values and the adjustment result to generate a projection image of the lane line group. Finally, the projection image is compared with the lane line image to obtain error information corresponding to the calibrated extrinsic parameter values. The lane line endpoints are imaged in the camera system based on the camera's true extrinsic parameter values, and the adjusted calibrated extrinsic parameter values are used to project the lane line endpoints in the image onto the ground plane. , generating a projection image corresponding to the lane line image, realizing independent quantification of the influence of each parameter error of the camera calibration extrinsic parameter value, not only intuitively showing the influence of the calibration extrinsic parameter value error on the lane line presentation effect, but also realizing independent quantitative analysis of each specific parameter error, thereby improving the accuracy and efficiency of the camera extrinsic parameter value calibration, and thus solving the technical problem in related technologies that it is difficult to reversely infer the calibration extrinsic parameter value error information from the lane line group indicator information due to the inability to quantitatively analyze the specific influence of the camera calibration extrinsic parameter value error information on the lane line group indicator information in the lane line bird's-eye view map.

[0067] Optionally, in step S10, acquiring a lane line image includes:

[0068] Step S101, obtaining lane line parameters and lane parameters, wherein the lane line parameters include lane line width, length, number and starting end position information, and the lane parameters include lane width, length, number and position information;

[0069] In step S101, the lane position information is used to represent the distance that the vehicle's center axis deviates from the lane centerline. By determining the position of the lane centerline, the positions of other lanes are determined.

[0070] Step S102: Generate a lane line image based on the lane line parameters and lane parameters.

[0071] Specifically, according to national standards, there are three standard patterns for the design of dashed and dashed lane lines, designated "242," "464," and "696," representing the combined lengths of solid and dashed segments: 2, 4, and 6 meters for solid lines, and 4, 6, and 9 meters for dashed lines, respectively. Generating a lane line group directly obtains the coordinates of the four most critical corner points of each lane line segment, eliminating the need to only obtain the true values of the lane lines when projecting a bird's-eye view.

[0072] For instance, Figure 2 FIG. 1 is a schematic diagram of a method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application. Figure 2As shown in the figure, the "696" model was used to design the virtual and real lane lines and used as the ground truth standard lane lines. The lane width was set to 0.15m, and the number of lane segments was set to 5 (counting from top to bottom). The lane width was set to 3.6m, and the number of lanes was set to 5. Each lane segment can be represented by 4 endpoints, so with 5 lane segments and 5 lanes, there are ultimately 4*5*(5+1)=120 endpoints.

[0073] Furthermore, to facilitate analysis and calibration, the starting point of the lane line is set at a position 10 meters longitudinally from the origin of the coordinate system. This setting ensures that the lane line starts from a point relatively far away from the camera, thus covering a wider field of view. In addition, the center axis of the middle lane is precisely aligned with the origin of the coordinate system, that is, the center axis of the vehicle deviates from the center line of the lane by 0 meters, which enhances the symmetry and balance of the model and facilitates a more accurate evaluation of the camera calibration effect. Table 2 shows the results of the calibration with the camera. Figure 2 The lane parameters and lane parameter settings corresponding to the lane line image shown are shown in Table 1. The lane line parameters include a lane width of 15 cm, a lane length of 600 cm, a lane segment count of 5, and a lane starting longitudinal distance of 10 m. The lane parameters include a lane width of 3.6 m, a lane length of 90 m, a lane count of 5, and a distance of 0 m between the vehicle's center axis and the lane centerline.

[0074] Table 2 Lane line parameters and lane parameter settings

[0075] Lane line parameters Lane parameters Lane width (cm): 15 Lane width (m): 3.6 Lane line length (cm): 600 Lane length (m): 90 Number of lane segments: 5 Number of lanes: 5 Lane line starting longitudinal distance (m): 10 Distance of vehicle center axis from lane centerline (m): 0

[0076] Based on the above steps S101 to S102, lane line parameters and lane parameters are obtained, and a lane line image is generated based on the lane line parameters and lane parameters. A lane line image that meets national standards, has high complexity and practicality, and contains rich lane line information is calculated. It not only conforms to the common layout of actual road traffic, enhances the practicality and authenticity of the solution, but also increases the complexity of the lane line group through its specific layout and size, providing more data points for subsequent analysis, which in turn helps to improve the accuracy of determining the error of the calibration external parameter value.

[0077] Optionally, in step S12, adjusting the calibration extrinsic parameter value of the camera based on the true extrinsic parameter value, and obtaining the adjustment result includes:

[0078] Step S121, obtaining the accuracy threshold of the calibration extrinsic parameter value;

[0079] Specifically, in the production line calibration environment, the accuracy of the three-dimensional coordinates of the feature points, the accuracy of feature detection, and the accuracy of the camera's intrinsic parameters can be guaranteed. Based on these prerequisites, it can be assumed that the accuracy of feature point recognition and camera intrinsic parameters has been effectively controlled during the calibration process. Therefore, the deviation between the calibrated extrinsic parameter values and the true extrinsic parameter values should be predictable and limited. In view of this, the deviation range of the three deflection angles (i.e., pitch angle, roll angle, and yaw angle) in the calibration extrinsic parameter values is set to within ±1 degree.

[0080] Step S122: adjusting at least one deflection angle parameter in the true extrinsic parameter value according to the accuracy threshold to obtain the deflection angle parameter in the adjustment result.

[0081] Specifically, to gain a deeper understanding of the impact of different deflection angle deviations on the calibration of extrinsic parameter values, the design of the calibration extrinsic parameter values was divided into three stages: First, the deviation of each rotation angle was adjusted individually. The independent changes in the pitch, roll, and yaw angles were studied separately to investigate how they affect lane group metrics. Second, the two deflection angle deviations were adjusted simultaneously. By simultaneously varying the pitch and roll angles, the pitch and yaw angles, and the roll and yaw angles, it was possible to evaluate how the two parameter deviations affect lane group metrics and whether this combined effect could be decomposed into its own independent influences, thereby determining the degree of coupling between the parameters. Finally, the three deflection angle deviations—pitch, roll, and yaw angles—were adjusted simultaneously to comprehensively evaluate the combined impact of the three parameter deviations on lane group metrics and the coupling of their influences.

[0082] For example, only one of the deflection angle parameters in the calibration extrinsic parameter values is adjusted, as shown in Tables 3A to 3C. In Table 3A, only the pitch angle is adjusted, in Table 3B, only the roll angle is adjusted, and in Table 3C, only the yaw angle is adjusted.

[0083] Table 3A Calibration parameter value settings

[0084] pitch(deg) -91 roll(deg) 0 yaw(deg) -90 Camera height (m) 1.5

[0085] Table 3B Calibration parameter value settings

[0086] pitch(deg) -90 roll(deg) 1 yaw(deg) -90 Camera height (m) 1.5

[0087] Table 3C Calibration parameter value settings

[0088] pitch(deg) -90 roll(deg) 0 yaw(deg) -91 Camera height (m) 1.5

[0089] For example, only two of the deflection angle parameters in the calibration extrinsic parameter values are adjusted, as shown in Tables 4A-4C. In Table 4A, only the pitch and roll angles are adjusted; in Table 4B, only the pitch and yaw angles are adjusted; and in Table 4C, only the roll and yaw angles are adjusted.

[0090] Table 4A Calibration parameter value settings

[0091] pitch(deg) -91 roll(deg) 1 yaw(deg) -90 Camera height (m) 1.5

[0092] Table 4B Calibration parameter value settings

[0093] pitch(deg) -91 roll(deg) 0 yaw(deg) -91 Camera height (m) 1.5

[0094] Table 4C Calibration parameter value settings

[0095] pitch(deg) -90 roll(deg) 1 yaw(deg) -91 Camera height (m) 1.5

[0096] For example, three deflection angle parameters in the calibration extrinsic parameter values are adjusted, as shown in Table 5. In Table 5, the pitch angle, roll angle, and yaw angle are adjusted.

[0097] Table 5 Calibration parameter value settings

[0098] pitch(deg) -91 roll(deg) 1 yaw(deg) -91 Camera height (m) 1.5

[0099] Based on the above steps S121 to S122, the accuracy threshold of the calibration extrinsic parameter value is obtained; at least one deflection angle parameter in the true value extrinsic parameter value is adjusted according to the accuracy threshold to obtain the deflection angle parameter in the adjustment result. This not only takes into account the accuracy range of the deflection angle, but also conducts a hierarchical analysis of the impact of the deviation, from a single parameter to a composite parameter, and then to the simultaneous action of three parameters. This can effectively separate the influence of different parameters, clarify the specific effect of each parameter on the lane line group indicator information, and after clarifying the degree of coupling between the parameters, it can be more targeted to adjust the parameters during calibration to avoid invalid or inefficient adjustments, thereby improving the overall efficiency and final accuracy of the calibration.

[0100] Optionally, the translation value in the adjustment result is set to the translation value of the true extrinsic parameter value.

[0101] Specifically, in a normal production line calibration environment, the 3D coordinate accuracy of feature points, feature point detection accuracy, and camera intrinsic parameter accuracy meet the requirements. Therefore, small deviations in the three translations of the calibration extrinsic parameter values have little impact on the projected lane line group and can be ignored. Therefore, the three translation values of the true extrinsic parameter values are directly used as the three translation values of the calibration extrinsic parameter values.

[0102] By taking into account that in the actual calibration process, small deviations in the three translation amounts have relatively little impact on the projected lane line group, this means that even if there is a slight translation error, the impact on the entire calibration effect may be masked or offset by other factors under the existing accuracy level. Therefore, the three translation amount values in the true value extrinsic parameter value are directly applied as the three translation amount values of the calibration extrinsic parameter value, which simplifies the camera calibration process and reduces the number of parameters that need to be adjusted, thereby avoiding additional adjustment or optimization of the translation amount in the calibration process, thereby reducing the complexity and uncertainty of the calibration process, and making the calibration process more efficient and controllable while ensuring calibration accuracy.

[0103] Optionally, in step S14, performing an imaging operation on the lane line group according to the true extrinsic parameter value and the adjustment result to generate a projection image of the lane line group includes:

[0104] Step S141, performing simulated imaging on the lane line group according to the true extrinsic parameter values to obtain an imaging result;

[0105] Step S142 : Projecting the imaging result onto a ground plane coordinate system using the adjustment result to generate a projection image of the lane line group.

[0106] Specifically, first, the geodetic coordinate system of the lane line group is converted to the camera's true coordinate system. Assume that the coordinates of each corner point in the standard lane line group (the four corner points of each lane line) in the geodetic coordinate system are (P_{earth}=(X_{earth},Y_{earth},Z_{earth})), and the camera's true external parameters include three rotation angles (i.e., pitch, roll, and yaw) and translations (lateral offset, longitudinal offset, and height offset). In order to achieve coordinate conversion, the rotation matrix (R_{true value}) and the translation matrix (T_{true value}) are applied together on (P_{earth}) to obtain the coordinates of the point in the camera's true coordinate system (P_{true value}). This process includes rotation and translation operations on the point, ensuring the consistency of the point's position information in different coordinate systems.

[0107] After converting to the true camera coordinate system, normalization is performed. This operation normalizes the z component of the coordinate value to 1. In other words, the coordinates (x, y, z) of the point are scaled and converted to (x / z, y / z, z / z), or (x / z, y / z, 1). Normalization is a crucial step in the coordinate conversion process, simplifying subsequent matrix operations and facilitating the projection of points from 3D space onto the image plane.

[0108] The intrinsic parameter matrix is then used to transform the normalized coordinates from the camera's true coordinate system to the image coordinate system. This process does not need to consider distortion factors because distortion is assumed to be non-existent. To transform the image coordinates back to the camera's calibration coordinate system, the inverse matrix of the intrinsic parameter matrix is directly used for the transformation. It is worth noting that both the transformation from the camera's true coordinate system to the image coordinate system and the inverse transformation from the image coordinates to the camera's calibration coordinate system always use the same set of intrinsic parameter values. This property means that the coordinates of the camera's true coordinate system and the camera's calibration coordinate system are numerically equivalent, simplifying the conversion process between coordinates.

[0109] Next, the camera calibration extrinsics are applied to perform a coordinate transformation. After converting the coordinates from the image coordinate system back to the calibration coordinate system, the camera calibration extrinsics are used to further transform the coordinates from the calibration coordinate system to the earth coordinate system. The earth coordinate values calculated in this step may not lie strictly on the earth plane (i.e., the z value is not 0) due to projection errors caused by deviations in the calibration extrinsics.

[0110] Finally, the geodetic plane coordinates are calculated. To accurately map the calculated geodetic coordinate system coordinates to the geodetic plane (i.e., z = 0), an additional coordinate transformation is required. This operation relies on the camera's calibrated extrinsic translation, which is set to be consistent with the true extrinsic translation in this process. The specific implementation method involves calculating the intersection of a spatial line with the geodetic plane, where the spatial line is determined by the camera's translation (Xo, Yo, Zo) and the geodetic coordinates (X1, Y1, Z1). To simplify the calculation, this process can be decomposed into two intersection calculations of a plane line with the z = 0 plane: first, the intersection of the plane line determined by (Xo, Zo) and (X1, Z1) with the z = 0 plane is calculated to obtain the x coordinate value on the geodetic plane; second, the intersection of the plane line determined by (Yo, Zo) and (Y1, Z1) with the z = 0 plane is calculated to obtain the y coordinate value on the geodetic plane.

[0111] Based on the above steps S141 to S142, the lane line group is simulated and imaged according to the true value extrinsic parameter values to obtain an imaging result; the imaging result is projected onto the ground plane coordinate system using the adjustment result to generate a projection image of the lane line group. By separating the effects of the true value extrinsic parameters and the calibration extrinsic parameters, high precision can be maintained during the simulation imaging process, and the direct impact of the calibration extrinsic parameter deviation on the lane line projection result is clearly demonstrated in the projection stage, which helps to accurately identify and isolate the source of the error, and the projection image generated by projecting onto the ground plane coordinate system intuitively shows the changes in the shape and position of the lane line group caused by the deviation of the calibration extrinsic parameter value, providing an intuitive basis for subsequent error analysis and correction.

[0112] Optionally, in step S16, the projected image is compared with the lane line image to obtain error information corresponding to the extrinsic parameter value, including:

[0113] Step S161: Visualize the projected image and the lane line image according to the horizontal pixel ratio and the vertical pixel ratio to obtain position information of the lane line group in the projected image and the lane line group in the lane line image in the same bird's-eye view;

[0114] In step S161, the horizontal and vertical pixel ratios are used to represent the actual horizontal distance corresponding to each pixel. The vertical and horizontal pixel ratios are used to represent the actual vertical distance corresponding to each pixel. The horizontal and vertical pixel ratios ensure that the projection image and lane marking image visualized in the bird's-eye view reflect the real-world scale, thereby enabling accurate comparison of the spatial positions of the two lane markings.

[0115] Specifically, we use a defined horizontal and vertical pixel ratio to adjust the lane marking groups in the projected image so that their positional information can be visually compared with that of the lane marking groups in the lane marking image within the same bird's-eye view. This operation not only visually presents the two lane marking groups in the same reference frame, but also allows for direct observation of their relative positional differences.

[0116] Exemplarily, the lane line group in the lane line image and the lane line group in the projection image are displayed on the bird's-eye view at a ratio of 1 pixel in the horizontal direction = 3 cm and 1 pixel in the vertical direction = 10 cm, visually displaying the positional relationship between the two lane line groups. Figure 3A is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 3A As shown in the figure, the light lane lines in the figure are the lane line group in the lane line image, and the dark lane lines in the figure are the lane line group in the projection image. The calibration extrinsic parameter values corresponding to the dark lane lines in the figure are shown in Table 3A. Figure 3B is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 3B As shown in the figure, the light lane lines are the lane line group in the lane line image, and the dark lane lines are the lane line group in the projected image. The calibration extrinsic parameter values corresponding to the dark lane lines in the figure are shown in Table 3B. Figure 3C is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 3C As shown in Figure 3C, the light lane lines in the figure are the lane line group in the lane line image, and the dark lane lines in the figure are the lane line group in the projected image. The extrinsic parameter values corresponding to the dark lane lines in the figure are shown in Table 3C.

[0117] Figure 4Ais a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 4A As shown in Figure 4A, the light lane lines in the figure are the lane line group in the lane line image, and the dark lane lines in the figure are the lane line group in the projected image. The extrinsic parameter values corresponding to the dark lane lines in the figure are shown in Table 4A. Figure 4B is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 4B As shown in Figure 4B, the light lane lines in the figure are the lane line group in the lane line image, and the dark lane lines in the figure are the lane line group in the projected image. The extrinsic parameter values corresponding to the dark lane lines in the figure are shown in Table 4B. Figure 4C is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 4C As shown in Figure 4, the light lane lines in the figure are the lane line group in the lane line image, and the dark lane lines in the figure are the lane line group in the projected image. The extrinsic parameter values corresponding to the dark lane lines in the figure are shown in Table 4C.

[0118] Figure 5 is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 5 As shown in the figure, the light lane lines are the lane line group in the lane line image, and the dark lane lines are the lane line group in the projection image. The calibration extrinsic parameter values corresponding to the dark lane lines in the figure are shown in Table 5.

[0119] Step S162: Determine error information corresponding to the calibration extrinsic parameter value based on the position information.

[0120] Based on the above steps S161 to S162, the projection image and the lane line image are visualized according to the horizontal pixel ratio and the vertical pixel ratio to obtain the position information of the lane line group in the projection image and the lane line group in the lane line image in the same bird's-eye view; the error information corresponding to the calibrated external parameter value is determined based on the position information, and the defined horizontal and vertical pixel ratios ensure that the visualization in the bird's-eye view can accurately reflect the actual size, thereby improving the accuracy of the error information determination, and by visually comparing the lane line groups in the bird's-eye view, the position differences of the lane line groups are intuitively displayed. By linking the visualization result display with the specific calibrated external parameter value, the source of the error can be quickly located, and the actual impact of the parameter deviation on the projection effect of the lane line group can be understood.

[0121] Optionally, in step S162, determining error information corresponding to the extrinsic parameter value based on the position information includes:

[0122] Step S201: determining lane line group index information based on the position information, wherein the lane line group index information includes lane line length and lane width in the lane line image in the bird's-eye view, distance information between the lane line and the camera in the projected image, and yaw angle information of the lane line;

[0123] In step S201, the length and width of the lane lines in the lane line image in the above-mentioned bird's-eye view include: the length of the far-end lane line, the length of the near-end lane line, the far-end lane width, and the near-end lane width in the lane line image in the bird's-eye view, wherein the near-end lane line length is the distance between the two longitudinal endpoints of the lane line segment of the nearest section, the far-end lane line length is the distance between the two longitudinal endpoints of the lane line segment of the farthest section, the near-end lane width is the distance between the two closest endpoints of the two lane line segments of the nearest section, and the far-end lane width is the distance between the two farthest endpoints of the two lane line segments of the farthest section.

[0124] For example, Figure 6 is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 6 As shown in the figure, the length of the far lane line, the length of the near lane line, the far lane width, and the near lane width in the lane line image in the bird's-eye view are displayed.

[0125] The distance information between the lane line and the camera in the projected image is used to represent the vertical distance from the camera to the lane lines on both sides. The yaw angle information of the lane line is used to represent the angle between the lane line and the vertical line.

[0126] For example, Figure 7 is a schematic diagram of another method for determining error information of camera calibration extrinsic parameter values provided by an embodiment of the present application, such as Figure 7 As shown in the figure, the vertical distance from the camera to the lane lines on both sides in the projected image in the bird's-eye view, as well as the angle between the lane lines and the vertical line are shown. Figure 7 The corresponding calibration parameter value settings are shown in Table 3A.

[0127] For example, Table 6 shows Figure 6 and Figure 7 The corresponding lane line group indicator information.

[0128] Table 6 Lane line group index information

[0129]

[0130] Step S202: Determine error information of the extrinsic parameter value based on the lane line group index information.

[0131] Specifically, based on the extracted lane group metrics (lane length, width, distance from the camera, and yaw angle), a comparative analysis is performed to convert the deviations in the camera's extrinsic calibration parameters into specific quantitative indicators, such as length deviation, position offset, and directional deviation. After quantifying the errors, the correlation between each indicator and the extrinsic calibration parameter values is further analyzed to identify which parameter deviations have a significant impact on the lane group metrics and whether these effects can be analyzed independently or are coupled. Based on these analysis results, a detailed error report is generated, describing how the deviations in the extrinsic calibration parameter values affect the various lane group metrics, providing clear guidance for subsequent parameter optimization and system adjustments.

[0132] Based on the above steps S201 to S202, the lane line group index information is determined based on the position information; the error information of the calibration external parameter value is determined based on the lane line group index information. By converting the position information into intuitive lane line group index information, not only the accurate quantification of the error of the camera calibration external parameter value is achieved, the interpretability of the data analysis and the operator's understanding of the results are enhanced, but also the specific impact of the error on the lane line group index and the interaction between the parameters are analyzed, thereby improving the accuracy and efficiency of error identification, providing quantitative data and analysis results for the optimization of camera calibration in the intelligent driving system, thereby helping to improve the reliability of the system.

[0133] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus the necessary general hardware platform, and of course it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0134] Example 2

[0135] The present application also provides a device 80 for determining error information of camera calibration external parameter values. Figure 8 , Figure 8It is a structural diagram of a device for determining error information of camera calibration extrinsic parameter values provided in an embodiment of the present application, including: an acquisition module 810, used to obtain the true extrinsic parameter value and lane line image of the camera, wherein the lane line image is used to determine the ground plane coordinates of the lane line group; an adjustment module 820, used to adjust the calibration extrinsic parameter value of the camera based on the true extrinsic parameter value to obtain an adjustment result; a generation module 830, used to perform an imaging operation on the lane line group according to the true extrinsic parameter value and the adjustment result to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging; a determination module 840, used to compare the projection image with the lane line image to obtain error information corresponding to the calibration extrinsic parameter value.

[0136] The above-mentioned camera calibration extrinsic parameter value error information determination device provided in the embodiment of the present application achieves the following technical effects: by imaging the lane line endpoints in the camera system according to the true extrinsic parameter values of the camera, and then using the adjusted calibration extrinsic parameter values to project the lane line endpoints in the image onto the ground plane, a projection image corresponding to the lane line image is generated, thereby achieving independent quantification of the influence of each parameter error of the camera calibration extrinsic parameter value, not only intuitively demonstrating the influence of the calibration extrinsic parameter value error on the lane line presentation effect, but also achieving independent quantitative analysis of each specific parameter error, thereby improving the accuracy and efficiency of the camera extrinsic parameter value calibration, and further solving the technical problem in the related art that it is difficult to reversely infer the calibration extrinsic parameter value error information from the lane line group indicator information due to the inability to quantitatively analyze the specific influence of the camera calibration extrinsic parameter value error information on the lane line group indicator information in the lane line bird's-eye view map.

[0137] It should be noted that the above modules can be implemented through software or hardware. For the latter, it can be implemented in the following ways, but not limited to: the above modules are all located in the same processor; or the above modules are located in different processors in any combination.

[0138] Example 3

[0139] The present application also provides an electronic device 90, please refer to Figure 9 , Figure 9 This is a structural diagram of an electronic device provided in one embodiment of the present application, including a processor 910 and a memory 920, wherein the memory 910 is used to store computer programs; the processor 920 is used to execute the program stored in the memory 910 to implement the method for determining the error information of the camera calibration external parameter value introduced in any embodiment of the present application.

[0140] Optionally, in this embodiment, the processor may be configured to execute the following steps through a computer program:

[0141] Step S1, obtaining the true extrinsic parameter values and lane line images of the camera, wherein the lane line images are used to determine the ground plane coordinates of the lane line group;

[0142] Step S2, adjusting the calibrated extrinsic parameter values of the camera based on the true extrinsic parameter values to obtain an adjustment result;

[0143] Step S3: performing an imaging operation on the lane line group based on the true extrinsic parameter values and the adjustment results to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging;

[0144] Step S4: Compare the projection image with the lane line image to obtain error information corresponding to the extrinsic parameter values.

[0145] The above-mentioned electronic device provided in the embodiment of the present application achieves the following technical effects: by imaging the lane line endpoints in the camera system according to the true value extrinsic parameter values of the camera, and then using the adjusted calibration extrinsic parameter values to project the lane line endpoints in the image onto the ground plane, a projection image corresponding to the lane line image is generated, thereby achieving independent quantification of the influence of each parameter error of the camera calibration extrinsic parameter value, not only intuitively demonstrating the influence of the calibration extrinsic parameter value error on the lane line presentation effect, but also achieving independent quantitative analysis of each specific parameter error, thereby improving the accuracy and efficiency of the camera extrinsic parameter value calibration, and further solving the technical problem in the related art that it is difficult to reversely infer the calibration extrinsic parameter value error information from the lane line group indicator information due to the inability to quantitatively analyze the specific influence of the camera calibration extrinsic parameter value error information on the lane line group indicator information in the lane line bird's-eye view.

[0146] It can be understood by those skilled in the art that Figure 9 The structure shown is for illustration only, and the electronic device may also be a terminal device such as a smart phone (such as an Android phone, an iOS phone, etc.), a tablet computer, a PDA, and a mobile Internet device (Mobile Internet Devices, MID for short). Figure 9 It does not limit the structure of the above electronic device. For example, the electronic device 90 may also include Figure 9 More or fewer components (such as network interfaces, display devices, etc.) shown in, or with Figure 9 Different configurations shown.

[0147] Example 4

[0148] An embodiment of the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, it implements the method for determining the error information of the camera calibration external parameter value introduced in any embodiment of the present application.

[0149] Optionally, in this embodiment, the storage medium may be configured to store a computer program for performing the following steps:

[0150] Step S1, obtaining the true extrinsic parameter values and lane line images of the camera, wherein the lane line images are used to determine the ground plane coordinates of the lane line group;

[0151] Step S2, adjusting the calibrated extrinsic parameter values of the camera based on the true extrinsic parameter values to obtain an adjustment result;

[0152] Step S3: performing an imaging operation on the lane line group based on the true extrinsic parameter values and the adjustment results to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging;

[0153] Step S4: Compare the projection image with the lane line image to obtain error information corresponding to the extrinsic parameter values.

[0154] Optionally, in this embodiment, the above-mentioned storage medium may include but is not limited to: a USB flash drive, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and other media that can store computer programs.

[0155] The above-mentioned electronic device provided in the embodiment of the present application achieves the following technical effects: by imaging the lane line endpoints in the camera system according to the true value extrinsic parameter values of the camera, and then using the adjusted calibration extrinsic parameter values to project the lane line endpoints in the image onto the ground plane, a projection image corresponding to the lane line image is generated, thereby achieving independent quantification of the influence of each parameter error of the camera calibration extrinsic parameter value, not only intuitively demonstrating the influence of the calibration extrinsic parameter value error on the lane line presentation effect, but also achieving independent quantitative analysis of each specific parameter error, thereby improving the accuracy and efficiency of the camera extrinsic parameter value calibration, and further solving the technical problem in the related art that it is difficult to reversely infer the calibration extrinsic parameter value error information from the lane line group indicator information due to the inability to quantitatively analyze the specific influence of the camera calibration extrinsic parameter value error information on the lane line group indicator information in the lane line bird's-eye view.

[0156] The serial numbers of the above embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0157] In the above embodiments of the present application, the description of each embodiment has its own focus. For parts that are not described in detail in a certain embodiment, please refer to the relevant description of other embodiments.

[0158] In this application, a plurality refers to two or more.

[0159] In this application, unless otherwise expressly defined, the terms "mounted," "connected," and "connected" should be interpreted broadly. For example, they can refer to fixed, detachable, or integral connections; mechanical or electrical connections; direct or indirect connections through an intermediary; and internal communication between two components. A person of ordinary skill in the art will understand the specific meanings of these terms in this application.

[0160] The terms "first," "second," "third," "fourth," etc. (if any) in this application are used to distinguish similar objects and are not necessarily used to describe a particular sequential order.

[0161] The term "and / or" in this application simply describes an association between related objects, indicating that three possible relationships exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. In addition, the character " / " in this application generally indicates that the related objects are in an "or" relationship.

[0162] Unless otherwise specified, all steps of the present application may be performed sequentially or randomly. For example, a statement that the method includes steps A and B indicates that the method may include steps A and B performed sequentially, or steps B and A performed sequentially. For example, a statement that the method may also include step C indicates that step C may be added to the method in any order, for example, the method may include steps A, B, and C, or steps A, C, and B, or steps C, A, and B, etc.

[0163] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements and improvements made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims

1. A method for determining error information of camera calibration extrinsic parameter values, characterized in that: include: Obtaining true extrinsic parameter values and a lane line image of a camera, wherein the lane line image is used to determine the ground plane coordinates of the lane line group; Adjusting the calibrated extrinsic parameter value of the camera based on the true extrinsic parameter value to obtain an adjustment result; performing an imaging operation on the lane line group according to the true extrinsic parameter value and the adjustment result to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging; The projection image is compared with the lane line image to obtain error information corresponding to the calibrated extrinsic parameter value.

2. The method according to claim 1, characterized in that Obtaining lane line images includes: Acquire lane line parameters and lane parameters, wherein the lane line parameters include lane line width, length, number and starting end position information, and the lane parameters include lane width, length, number and position information; The lane line image is generated based on the lane line parameter and the lane parameter.

3. The method according to claim 1, characterized in that Adjusting the calibrated extrinsic parameter value of the camera based on the true extrinsic parameter value to obtain the adjustment result includes: Obtaining a precision threshold of the extrinsic parameter value; At least one deflection angle parameter in the true value extrinsic parameter value is adjusted according to the accuracy threshold to obtain the deflection angle parameter in the adjustment result.

4. The method according to claim 3, characterized in that The translation value in the adjustment result is set as the translation value of the true value extrinsic parameter value.

5. The method according to claim 1, characterized in that Performing an imaging operation on the lane line group according to the true extrinsic parameter value and the adjustment result to generate a projection image of the lane line group includes: Performing simulated imaging on the lane line group according to the true extrinsic parameter value to obtain an imaging result; The imaging result is projected onto a ground plane coordinate system using the adjustment result to generate a projection image of the lane line group.

6. The method according to claim 1, characterized in that Comparing the projection image with the lane line image to obtain error information corresponding to the extrinsic parameter value includes: Visualizing the projected image and the lane line image according to a horizontal pixel ratio and a vertical pixel ratio to obtain position information of the lane line group in the projected image and the lane line group in the lane line image in the same bird's-eye view; Error information corresponding to the extrinsic parameter value is determined based on the position information.

7. The method according to claim 6, characterized in that Determining error information corresponding to the extrinsic parameter value based on the position information includes: Determining lane line group indicator information based on the position information, wherein the lane line group indicator information includes lane line length and lane width in the lane line image in the bird's-eye view, distance information between the lane line and the camera in the projected image, and yaw angle information of the lane line; Determine error information of the calibrated extrinsic parameter value based on the lane line group index information.

8. A device for determining error information of camera calibration extrinsic parameter values, characterized in that: include: an acquisition module, configured to acquire true extrinsic parameter values and a lane line image from a camera, wherein the lane line image is used to determine the ground plane coordinates of the lane line group; An adjustment module, configured to adjust a calibration extrinsic parameter value of the camera based on the true extrinsic parameter value to obtain an adjustment result; a generating module, configured to perform an imaging operation on the lane line group based on the true extrinsic parameter value and the adjustment result, to generate a projection image of the lane line group, wherein the projection image is used to determine the ground plane coordinates of the lane line group after imaging; The determination module is used to compare the projection image with the lane line image to obtain error information corresponding to the calibration extrinsic parameter value.

9. An electronic device, characterized in that: include: processor; a memory for storing instructions executable by the processor; The processor is configured to execute the instructions to implement the method for determining the error information of the camera calibration external parameter value according to any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, wherein the computer program is configured to execute the method for determining error information of camera calibration extrinsic parameter values as described in any one of claims 1 to 7 when running.

Images