An Automatic Calibration Method, Device, Equipment and Medium for the Height of a Vehicle-mounted Camera

Abstract

The present invention provides an automatic calibration method, device, equipment and medium for the height of an in-vehicle camera. The method includes: S1. Obtain the actual distance interval Δd between two end points A and B of the blank segment in the dotted lane line; and in the image coordinate system of the in-vehicle camera, calculate the ordinates a y , b y of the imaging points a and b of the two end points A and B in the image coordinate system; S2. Substitute a y , b y into the formula derived from the actual distance interval Δd between two end points A and B of the blank segment in the dotted lane line to calculate the height h of the in-vehicle camera. By locating the solid and dotted parts of the standard lane line and combining with the distance estimation formula, the present invention automatically calculates the camera height, reduces the manual calibration error and cost, and enhances the robustness of the calibration parameters.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and particularly to an automatic calibration method, device, equipment and medium for the height of an in-vehicle camera. Background Art

[0002] With the development of deep learning technology, especially the application of convolutional neural networks in the field of images. Advanced Driving Assistance System (ADAS) based on lens monitoring and pure images, as well as autonomous driving systems, have developed rapidly. Whether it is a family car, or various operating vehicles such as buses, coaches, freight vehicles, and muck trucks that are regulated by the state, intelligent auxiliary driving systems are being continuously installed to assist in driving safety. For example, in the ADAS system, the mainstream is to install a single-eye or binocular camera (usually installed on the windshield), and some also install sensor devices such as radar and lidar. These cameras, radars, etc. are used to monitor the driving state of the vehicle. When potential dangers occur during vehicle driving, such as lane deviation, too close vehicle distance, or possible collision with the vehicle in front, an alarm reminder is sent in time, which can help the driver reduce the occurrence of traffic accidents and assist the driver in driving safely. Further, for example, in the autonomous driving system, the functions of assisting or replacing the driver to operate the vehicle are realized through perception, positioning, decision-making and control algorithms, which can prevent and reduce driving accidents caused by human operation errors. At the same time, the intelligent system can lower the driving threshold, and basically ensure that each driver can make correct decisions through the assistance system. The most basic and core prerequisite in the autonomous driving system is to be able to sense the surrounding objects and calculate the distance between the vehicle itself and each object through a formula, providing the most basic and reliable data basis for subsequent decision-making.

[0003] Overall, whether it is an ADAS system or an autonomous driving system, if a monocular camera is used, distance estimation is a fundamental and core part. Only by ensuring the correctness of distance calculation can the subsequent decision-making and control of the system be effective. In the distance estimation algorithm, among all the parameters that the camera needs to be calibrated, the height of the camera installation position from the ground is a very crucial parameter. Considering the impact of calibration parameter accuracy on distance estimation error, obtaining the correct camera height becomes the basic guarantee for distance estimation accuracy. However, in the actual use process, for the parameter of camera height, in most scenarios, it needs to be manually calibrated, and manual calibration inevitably brings human errors, and due to different calibration scenarios, the resulting manual calibration costs are also difficult to calculate. In addition, it cannot be excluded that the camera is accidentally touched by humans during use or the camera attitude changes due to unpredictable external factors such as vehicle driving bumps, resulting in a large distance estimation error in autonomous driving. If only a single manual calibration is performed after the camera is installed, it is inevitable that the robustness of the system after long-term operation cannot be guaranteed. If manual correction is required later, it will increase the later maintenance costs.

[0004] In addition to the above intelligent driving systems, most application scenarios based on camera distance estimation require calibrating the camera height. However, due to the limitations of the use scenarios, it is often necessary to capture the marker features of certain specific scenarios when the camera is stationary and calculate the camera height using specific trigonometric relationships. For in-vehicle cameras that follow vehicles and drive on the road for a long time, this static estimation is not reasonable, so it is necessary to find more common markers and formulas to achieve dynamic automatic calibration of the camera. Summary of the Invention

[0005] The technical problem to be solved by the present invention is to provide an automatic calibration method, device, equipment and medium for the height of an in-vehicle camera. By locating the solid and dashed parts of the standard lane line and combining the distance estimation formula, the camera height is automatically calculated, reducing manual calibration errors and costs and enhancing the robustness of the calibration parameters.

[0006] In the first aspect, the present invention provides an automatic calibration method for the height of an in-vehicle camera, including the following steps:

[0007] S1. Obtain the actual distance interval Δd between the two endpoints A and B of the blank segment in the dashed lane line; and in the image coordinate system of the in-vehicle camera, calculate the vertical coordinates a y 、b y of the imaging points a and b of the two endpoints A and B in the image coordinate system;

[0008] S2. Substitute a y 、b y into the following formula to calculate the height h of the in-vehicle camera:

[0009]

[0010] wherein, d y is the actual size of a single pixel, and f is the focal length of the vehicle-mounted camera; g y is the ordinate of the optical center point g in the image coordinate system; v y is the ordinate of the imaging point v of the horizon vanishing point in the image coordinate system.

[0011] In a second aspect, the present invention provides an automatic calibration device for the height of a vehicle-mounted camera, including:

[0012] A lane line parameter acquisition module, configured to obtain the actual distance interval Δd between two end points A and B of the blank segment in the dotted lane line;

[0013] An ordinate calculation module, configured to calculate, in the image coordinate system of the vehicle-mounted camera, the average ordinate a y of point a and the average ordinate b y of point b through two imaging points a and b of the two end points A and B on the imaging plane;

[0014] A vehicle-mounted camera height calculation module, configured to substitute a y , b y into the following formula to calculate the height h of the vehicle-mounted camera:

[0015]

[0016] wherein, d y is the actual size of a single pixel, and f is the focal length of the vehicle-mounted camera; g y is the ordinate of the optical center point g in the image coordinate system; v y is the ordinate of the imaging point v of the horizon vanishing point in the image coordinate system.

[0017] In a third aspect, the present invention provides an electronic device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, wherein when the processor executes the program, the method described in the first aspect is implemented.

[0018] In a fourth aspect, the present invention provides a computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, the method described in the first aspect is implemented.

[0019] One or more technical solutions provided in the embodiments of the present invention have at least the following technical effects or advantages: The present invention uses a lane line detection algorithm (including but not limited to traditional algorithms or deep learning methods, etc.) to locate the lane lines in the image, and obtains a pair of pixel coordinates of the stable and reliable ends of the dotted line in the image through constraint conditions; According to the derivation of the monocular distance estimation formula, the two-point coordinates and the actual distance of the two ends of the obtained dotted line are brought into the monocular distance estimation formula, so as to obtain the current camera height; Since the lane lines are commonly visible on the road, during the driving process of the vehicle, the automatically calibrated camera height realizes dynamic estimation, which is more in line with the actual use scenario than static estimation, and greatly improves the feasibility of the solution.

[0020] The above description is only an overview of the technical solution of the present invention. In order to be able to understand the technical means of the present invention more clearly, it can be implemented in accordance with the content of the description. And in order to make the above and other purposes, features and advantages of the present invention more obvious and understandable, the following specifically illustrates the specific embodiments of the present invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] The present invention will be further described below with reference to the accompanying drawings in conjunction with embodiments.

[0022] Figure 1 It is a flowchart of the method in the first embodiment of the present invention;

[0023] Figure 2 It is a schematic diagram of the imaging plane of the vehicle-mounted camera in the embodiment of the present invention;

[0024] Figure 3 It is a side view of the image coordinate system of the vehicle-mounted camera in the embodiment of the present invention;

[0025] Figure 4 It is the other side view of the image coordinate system of the vehicle-mounted camera in the embodiment of the present invention;

[0026] Figure 5 It is a schematic diagram of the structure of the device in the second embodiment of the present invention;

[0027] Figure 6 It is a schematic diagram of the structure of the electronic device in the third embodiment of the present invention;

[0028] Figure 7 It is a schematic diagram of the structure of the medium in the fourth embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0029] The embodiments of the present application provide an automatic calibration method, device, equipment and medium for the height of a vehicle-mounted camera. By locating the solid and dotted parts of the standard lane lines and combining the distance estimation formula, the camera height is automatically calculated, reducing the manual calibration error and cost, and enhancing the robustness of the calibration parameters.

[0030] The overall idea of the technical solution in the embodiments of this application is as follows: Use a lane line detection algorithm (including but not limited to traditional algorithms or deep learning methods, etc.) to locate the lane lines in the image, and obtain the pixel coordinate pairs of the two ends of the stable and reliable dotted line in the image through constraint conditions; According to the derivation of the monocular distance estimation formula, substitute the two-point coordinates and the actual distance of the two ends of the obtained dotted line into the monocular distance estimation formula to obtain the current camera height; Since lane lines are commonly visible on roads, during the driving of the vehicle, the automatically calibrated camera height realizes dynamic estimation, which is more in line with the actual usage scenario than static estimation, greatly improving the feasibility of the solution.

[0031] Embodiment 1

[0032] As Figure 1 shown, this embodiment provides an automatic calibration method for the height of an in-vehicle camera, including the following steps:

[0033] S1. Obtain the actual distance interval Δd between the two endpoints A and B of the blank segment in the lane line dotted line; and in the image coordinate system of the in-vehicle camera, calculate the vertical coordinates a y 、b y of the imaging points a and b of the two endpoints A and B in the image coordinate system;

[0034] S2. Substitute a y 、b y into the following formula to calculate the height h of the in-vehicle camera:

[0035]

[0036] In the formula, d y is the actual size of a single pixel, f is the focal length of the in-vehicle camera; g y is the vertical coordinate of the optical center point g in the image coordinate system; v y is the vertical coordinate of the imaging point v of the horizon vanishing point in the image coordinate system, and these parameter values can all be obtained through camera calibration.

[0037] As Figure 2 shown, in step S1, if the imaging points of the two endpoints A and B of the blank segment in the lane line dotted line in the image coordinate system are two points a and b, then the calculation process of the vertical coordinates a y and b y is specifically as follows:

[0038] S11. In the image coordinate system of the in-vehicle camera, locate the vertical coordinates of points a and b through a traditional algorithm or a deep learning algorithm to obtain the candidate coordinates of points a and b;

[0039] S12. Screen the candidate coordinates of points a and b, and only select the coordinates when the slopes of the left and right lane lines are close to obtain a screened coordinate set;

[0040] S13. Take the average of multiple frames of the screened coordinate set to obtain the average vertical coordinates a y and b y .

[0041] In the step S2, the derivation process of the height h of the vehicle-mounted camera is as follows:

[0042] Establish an image coordinate system of the vehicle-mounted camera. In the side view of this image coordinate system, point O c represents the vehicle-mounted camera, point I is the orthographic projection of point O c on the ground, O c G is the optical axis of the vehicle-mounted camera, and g is the optical center point where O c G intersects perpendicularly with the imaging plane. Then the focal length f of the vehicle-mounted camera is f = O C g; O c Q is the lower boundary of the camera imaging, and q is the near point where O c Q intersects with the imaging plane; O c V is parallel to the ground, V is the vanishing point of the horizon, and v is the intersection point of O c V and the imaging plane. P is the lower edge of the front vehicle's frame; point a is the intersection point of O c A and the imaging plane, and point b is the intersection point of O c B and the imaging plane. d2 is the distance from point I to point P;

[0043] Then first derive the height formula of the camera based on d2 using the monocular distance estimation formula, that is:

[0044]

[0045] In formula 1, α = ∠O C PI = ∠VO C P = ∠VO C G - ∠PO C G;

[0046] Then

[0047] Then

[0048]

[0049] Then

[0050] As can be seen from the above formula 6, when the internal parameters of the camera and the vanishing point of the horizon are known, the height h of the camera is only related to the horizontal distance d2 between the lower edge P of the front vehicle's frame and the camera, and the pixel coordinate p of the lower edge of the front vehicle's frame in the imaging plane y Therefore, if d2 and p are known y then the height h of the camera can be directly calculated, and the calculation formula is as follows:

[0051]

[0052] As Figure 4 shown, according to the national road regulations, the size of the lane line dividing line is usually a fixed value. The distance interval Δd between the two end points A and B of the blank section in the dotted line of the lane line in our country is 9 meters. According to this prior knowledge and the vertical coordinates a y and b y , three equations about h can be established according to formula 7, namely formula 8 to formula 10:

[0053]

[0054] d b = d a +Δd Formula 10

[0055] Then, combining formula 8, formula 9 and formula 10, we get:

[0056]

[0057] Combining formula 8 and formula 11 again, we get the final calculation formula for the height h of the vehicle-mounted camera:

[0058]

[0059] According to formula 1, by directly substituting the vertical coordinates a y and b y of the imaging points a and b of the two end points A and B obtained into it, the height h of the camera can be obtained; similarly, since this method is inspired by formula 6, as long as an estimation formula establishing the corresponding relationship between the actual distance d and the height h of the camera is established, this method can be extended and used. By using the prior knowledge of the lane line dotted line interval, the automatic calibration of the camera height is realized.

[0060] Based on the same inventive concept, the present application also provides a device corresponding to the method in Embodiment 1. For details, see Embodiment 2.

[0061] Embodiment 2

[0062] As Figure 5 shown, in this embodiment, an automatic calibration device for the height of a vehicle-mounted camera is provided, including:

[0063] The lane line parameter acquisition module is used to obtain the actual distance interval Δd between the two endpoints A and B of the blank section in the lane line dashes.

[0064] The vertical coordinate calculation module is used to calculate the average vertical coordinate a of point a in the image coordinate system of the vehicle-mounted camera through the two imaging points a and b of the two endpoints A and B on the imaging plane. y and the average vertical coordinate b of point b y ;

[0065] The vehicle-mounted camera height calculation module is used to substitute a y , b y into the following formula to calculate the height h of the vehicle-mounted camera:

[0066]

[0067] In the formula, d y is the actual size of a single pixel, and f is the focal length of the vehicle-mounted camera; g y is the vertical coordinate of the optical center point g in the image coordinate system; v y is the vertical coordinate of the imaging point v of the horizon vanishing point in the image coordinate system. These parameter values can all be obtained through camera calibration.

[0068] As Figure 2 shown, the process of the vertical coordinate calculation module calculating the average vertical coordinates a y and b y is specifically as follows:

[0069] S11. In the image coordinate system of the vehicle-mounted camera, locate the vertical coordinates of points a and b through traditional algorithms or deep learning algorithms to obtain the candidate coordinates of points a and b;

[0070] S12. Screen the candidate coordinates of points a and b, and only select the coordinates when the slopes of the left and right lane lines are close to obtain the screened coordinate set;

[0071] S13. Take the average of multiple frames of the screened coordinate set to obtain the average vertical coordinates a y and b y .

[0072] The specific derivation process of the formula for the height h of the vehicle-mounted camera is as follows:

[0073] Establish the image coordinate system of the vehicle-mounted camera. In the side view of this image coordinate system, point O c represents the vehicle-mounted camera, point I is the orthographic projection of point O c on the ground, O c G is the optical axis of the vehicle-mounted camera, and g is O cThe optical center point G that intersects the imaging plane perpendicularly, then the focal length f of the vehicle-mounted camera = O C g; O c Q is the lower boundary of the camera's imaging, and q is O c The near point where Q intersects the imaging plane; O c V is parallel to the ground, V is the vanishing point of the horizon, and v is O c The intersection point of V and the imaging plane, P is the lower edge of the front vehicle's frame; point a is O c The intersection point of A and the imaging plane, point b is O c The intersection point of B and the imaging plane, d2 is the distance from point I to point P;

[0074] Then, based on d2, use the monocular distance estimation formula to derive the camera height formula, that is:

[0075]

[0076] In formula 1, α = ∠O C PI = ∠VO C P = ∠VO C G - ∠PO C G;

[0077] Then

[0078]

[0079] Then

[0080] As can be seen from the above formula 6, when the camera internal parameters and the vanishing point of the horizon are known, the camera height h only depends on the horizontal distance d2 between the lower edge P of the front vehicle's frame and the camera and the pixel coordinate p of the lower edge of the front vehicle's frame in the imaging plane y Therefore, if d2 and p are known y Then the camera height h can be directly calculated, and the calculation formula is as follows:

[0081]

[0082] Such as Figure 4 shown, according to the national road regulations, the size of the lane line demarcation is usually a fixed value. The distance interval Δd between the two end points A and B of the blank section in the dotted line of the lane line in our country is 9 meters. Based on this prior knowledge and the vertical coordinates a y and b y , three equations about h can be established according to formula 7, that is, formulas 8 to 10:

[0083]

[0084] d b = da +Δd, Formula 10

[0085] Combining Formula 8, Formula 9, and Formula 10, we get:

[0086]

[0087] Combining Formula 8 and Formula 11 again, we obtain the final calculation formula for the height h of the vehicle-mounted camera:

[0088]

[0089] According to Formula 1, by directly substituting into it, the ordinates a y and b y of the imaging points a and b of the two endpoints A and B can be obtained, and then the camera height h can be obtained; similarly, since this method is inspired by Formula 6, as long as the distance estimation formula that establishes the corresponding relationship between the actual distance d and the camera height h is available, this method can be extended and used. By utilizing the prior knowledge of the dashed line interval of the lane lines, the automatic calibration of the camera height can be realized.

[0090] Since the device introduced in the second embodiment of the present invention is the device adopted for implementing the method of the first embodiment of the present invention, based on the method introduced in the first embodiment of the present invention, those skilled in the art can understand the specific structure and variations of the device, so it will not be elaborated here. Any device adopted for the method of the first embodiment of the present invention falls within the scope of protection of the present invention.

[0091] Based on the same inventive concept, this application provides an electronic device embodiment corresponding to the first embodiment, as detailed in the third embodiment.

[0092] Embodiment 3

[0093] This embodiment provides an electronic device, as Figure 6 shown, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the computer program, any implementation manner in the first embodiment can be realized.

[0094] Since the electronic device introduced in this embodiment is the device adopted for implementing the method in the first embodiment of this application, based on the method introduced in the first embodiment of this application, those skilled in the art can understand the specific implementation manner and various variations of the electronic device in this embodiment. Therefore, how the electronic device realizes the method in the embodiments of this application will not be introduced in detail here. Any device adopted by those skilled in the art for implementing the method in the embodiments of this application falls within the scope of protection of this application.

[0095] Based on the same inventive concept, this application provides a storage medium corresponding to the first embodiment, as detailed in the fourth embodiment.

[0096] Example 4

[0097] This embodiment provides a computer-readable storage medium, as Figure 7 shown, on which a computer program is stored. When the computer program is executed by a processor, any implementation manner in Embodiment 1 can be implemented.

[0098] The technical solution provided in the embodiments of the present application has at least the following technical effects or advantages: The lane line detection algorithm (including but not limited to traditional algorithms or deep learning methods, etc.) is used to locate the lane lines in the image, and the pixel coordinate pairs of the two ends of the stable and reliable dotted line in the image are obtained through constraint conditions; According to the derivation of the monocular distance estimation formula, the two-point coordinates and the actual distance of the two ends of the dotted line are brought into the monocular distance estimation formula, so as to obtain the current camera height; Since the lane lines are commonly visible on the road, during the driving of the vehicle, the automatically calibrated camera height realizes dynamic estimation, which is more in line with the actual use scenario than static estimation, and greatly improves the feasibility of the solution.

[0099] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, apparatus or system, or a computer program product. Therefore, the present invention can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0100] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0101] These computer program instructions can also be stored in a computer-readable memory capable of guiding a computer or other programmable data processing devices to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the functions in Figure 1 one process or multiple processes and / or blocks Figure 1The functions specified in one or more boxes.

[0102] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the steps of the functions specified in one or more processes and / or boxes Figure 1 One process or more processes and / or boxes Figure 1 The steps of the functions specified in one or more boxes.

[0103] Although the specific embodiments of the present invention have been described above, those skilled in the art of this technology should understand that the specific embodiments we described are illustrative rather than used to limit the scope of the present invention. Equivalent modifications and variations made by those skilled in the art in accordance with the spirit of the present invention should be covered by the scope protected by the claims of the present invention.

Claims

1. An automatic calibration method for the height of a vehicle-mounted camera, characterized in that: Including the following steps: S1. Obtain the actual distance interval Δd between the two end points A and B of the blank segment in the lane line dashes; and in the image coordinate system of the vehicle-mounted camera, calculate the ordinates a y , b y ; S2. Substitute a y and b y into the following formula to calculate the height h of the vehicle-mounted camera: where d y is the actual size of a single pixel, and f is the focal length of the vehicle-mounted camera; g y is the ordinate of the optical center point g in the image coordinate system; v y is the ordinate of the imaging point v of the horizon vanishing point in the image coordinate system.

2. The method according to claim 1, wherein: The ordinate a of the step S1 y and b y The specific calculation process is as follows: S11. In the image coordinate system of the vehicle-mounted camera, locate the ordinates of points a and b through a deep learning algorithm to obtain the candidate coordinates of points a and b; S12. Screen the candidate coordinates of points a and b, and only select the coordinates when the slopes of the left and right lane lines are close to obtain the screened coordinate set; S13. Average the coordinates in the filtered coordinate set over multiple frames to obtain the average vertical coordinates a y and b y .

3. The method according to claim 1, characterized in that: In step S2, the specific derivation process of the height h of the vehicle-mounted camera is as follows: Establish an image coordinate system for the in-vehicle camera. In the side view of this image coordinate system, point O c represents the in-vehicle camera, and point I is the orthographic projection of point O c on the ground. O c G is the optical axis of the in-vehicle camera, and g is the optical center point where O c G intersects perpendicularly with the imaging plane. Then the focal length f of the in-vehicle camera is f = O C g; O c Q is the lower boundary of the camera's imaging, and q is the near point where O c Q intersects with the imaging plane; O c V is parallel to the ground. V is the vanishing point of the horizon, and v is the intersection point of O c V and the imaging plane. P is the lower edge of the front vehicle's frame; point a is the intersection point of O c A and the imaging plane, and point b is the intersection point of O c B and the imaging plane. d2 is the distance from point I to point P; First, based on d2, derive the height formula of the camera using the following formula, that is: Then, based on the actual distance interval Δd between the two endpoints A and B of the blank segment in the lane line dotted line, derive the height formula of the camera, and it can be obtained: d b = d a + Δd Equation 10 Then, combining formula 8, formula 9, and formula 10, we get: Combining formula 8 and formula 11 again to obtain the final calculation formula for the height h of the vehicle-mounted camera:

4. The method according to claim 1, wherein: The actual distance interval Δd between the two endpoints A and B of the blank segment in the lane line dotted line is obtained from the dimensions of the lane line demarcation line in the national road regulations.

5. An automatic calibration device for the height of a vehicle-mounted camera, characterized in that: Including: A lane line parameter acquisition module for obtaining the actual distance interval Δd between the two endpoints A and B of the blank segment in the lane line dotted line; The vertical coordinate calculation module is used to calculate the average vertical coordinate a of point a in the image coordinate system of the vehicle-mounted camera through the two imaging points a and b of the two end points A and B on the imaging plane y and the average vertical coordinate b of point b y ; In-vehicle camera height calculation module, which is used to substitute a y , b y into the following formula to calculate the height h of the in-vehicle camera: where d y is the actual size of a single pixel, and f is the focal length of the vehicle-mounted camera; g y is the ordinate of the optical center point g in the image coordinate system; v y is the ordinate of the imaging point v of the horizon vanishing point in the image coordinate system.

6. The device according to claim 5, characterized in that: The ordinate calculation module calculates the average ordinates a y and b y The specific process is as follows: S11. In the image coordinate system of the vehicle-mounted camera, locate the ordinates of points a and b through a deep learning algorithm to obtain the candidate coordinates of points a and b; S12. Screen the candidate coordinates of points a and b, and only select the coordinates when the slopes of the left and right lane lines are close to obtain the screened coordinate set; S13. Perform multi-frame averaging on the filtered coordinate set to obtain the average vertical coordinates a y and b y .

7. The device according to claim 5, characterized in that: The specific derivation process of the height h formula of the vehicle-mounted camera is as follows: Establish an image coordinate system for the in-vehicle camera. In the side view of this image coordinate system, point O c represents the in-vehicle camera, and point I is the orthographic projection of point O c on the ground. O c G is the optical axis of the in-vehicle camera, and g is the optical center point where O c G intersects perpendicularly with the imaging plane. Then the focal length f of the in-vehicle camera is f = O C g; O c Q is the lower boundary of the camera's imaging, and q is the near point where O c Q intersects with the imaging plane; O c V is parallel to the ground. V is the vanishing point of the horizon, and v is the intersection point of O c V and the imaging plane. P is the lower edge of the front vehicle's frame; point a is the intersection point of O c A and the imaging plane, and point b is the intersection point of O c B and the imaging plane. d2 is the distance from point I to point P; First, based on d2, derive the height formula of the camera using the following formula, that is: Then, based on the actual distance interval Δd between the two endpoints A and B of the blank segment in the lane line dotted line, derive the height formula of the camera, and it can be obtained: d b = d a + Δd Equation 10 Then, combining formula 8, formula 9, and formula 10, we get: Combining formula 8 and formula 11 again to obtain the final calculation formula for the height h of the vehicle-mounted camera:

8. The device according to claim 5, characterized in that: The lane line parameter acquisition module obtains the actual distance interval Δd between the two endpoints A and B of the blank segment in the lane line dotted line by querying the dimensions of the lane line demarcation line in the national road regulations.

9. An electronic device, comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the method described in any one of claims 1 to 4.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by the processor, it implements the method described in any one of claims 1 to 4.

Images