Ramp line parking space position determination method and system based on visual correction and vehicle

Abstract

The invention discloses a method and system for determining the position of a ramp line parking space based on visual correction and a vehicle, and relates to the technical field of intelligent vehicles, and the method comprises the steps: obtaining an original image of the ramp line parking space through a vehicle-mounted camera, recognizing the ramp line parking space in the image through an image recognition algorithm, and determining the angular point information of the ramp line parking space in the original image; the gradient difference between the plane where the current vehicle is located and the plane where the linear parking space is located is obtained, and if the gradient difference is equal to 0, a virtual parking space line is directly projected according to the angular point information of the linear parking space; if the gradient difference is not equal to 0, correcting the angular point information of the linear parking space, and projecting the linear parking space into a virtual parking space line according to the corrected angular point information; and determining the real position of the line parking space on the plane according to the virtual parking space line. When the vehicle and the parking space are not in the same horizontal plane, error correction is performed on ramp line parking space identification based on a visual identification algorithm, so that the accurate position of the ramp line parking space in the plane is determined, and the parking success rate in a slope scene is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent vehicles, and particularly to a method, a system and a vehicle for determining the position of a ramp line parking space based on vision correction. Background Art

[0002] As an important part of vehicle intelligent driving technology, the automatic parking assistance technology has strong market demand. Currently, the mainstream automatic parking assistance technologies mostly use ultrasonic radars and in-vehicle cameras to detect and identify parking spaces, and then perform automatic parking after accurately identifying the parking spaces. Therefore, accurately identifying the position of the parking space is an important prerequisite for the automatic parking assistance technology.

[0003] However, the existing vision-based parking space recognition methods rely heavily on the environment of the target parking space. When the line parking space is in a sloped scenario, due to the slope, the accuracy of vision recognition is poor, and there is a large error between the recognized position of the line parking space and the actual position, which easily leads to problems such as incorrect parking postures and multiple parking attempts, resulting in a low final parking success rate and long time consumption. Summary of the Invention

[0004] To solve the deficiencies of the above-mentioned existing technologies, the present invention provides a method, a system and a vehicle for determining the position of a ramp line parking space based on vision correction. When the vehicle and the parking space are not on the same ground plane or horizontal plane, a vision recognition algorithm is used to recognize the ramp line parking space, and combined with the slope difference, the recognized ramp line parking space is corrected for errors, so as to accurately determine the position on the plane where the ramp line parking space is located, which can greatly improve the parking success rate in the sloped scenario and reduce the time consumption of automatic parking.

[0005] In the first aspect, the present invention provides a method for determining the position of a ramp line parking space based on vision correction.

[0006] A method for determining the position of a ramp line parking space based on vision correction includes:

[0007] Obtain the original image of the line parking space by using an in-vehicle camera, and identify the line parking space in the image through an image recognition algorithm to determine the corner point information of the line parking space in the original image; wherein, the corner point information includes the position information of the four corner points of the line parking space;

[0008] Obtain the slope difference between the plane where the current vehicle is located and the plane where the line parking space is located. If the slope difference is equal to 0, directly project the corner point information of the line parking space into virtual parking lines; if the slope difference is not equal to 0, correct the corner point information of the line parking space, and then project the corrected corner point information into virtual parking lines;

[0009] Determine the true position of the line parking space on the plane where it is located according to the virtual parking lines.

[0010] A further technical solution is that after obtaining the corner point information of the line parking space in the original image, preprocess the corner point information, including:

[0011] Abnormal point elimination: For all identified pixel points, calculate the distance of each pixel point from the pixel-level corner points of the line parking space obtained by fitting, and eliminate the pixel points whose distance exceeds the set distance.

[0012] Re-fitting: Perform quadratic fitting based on the eliminated pixel points and output accurate corner point information of the line parking space.

[0013] A further technical solution is to correct the corner point information of the line parking space and then project virtual parking space lines based on the corrected corner point information, including:

[0014] Project the corner point information of the line parking space in the original image onto the plane where the current vehicle is located to obtain virtual corner point information with errors; among them, the plane where the original image is located is always perpendicular to the plane where the current vehicle is located.

[0015] According to the installation height of the on-vehicle camera, the slope difference between the plane where the current vehicle is located and the plane where the line parking space is located, and the virtual corner point information with errors, substitute them into a preset correction relational expression to output the corrected virtual corner point information.

[0016] Determine the virtual parking space lines projected onto the plane where the line parking space is located based on the corrected virtual corner point information.

[0017] A further technical solution is that the determination of the correction relational expression includes:

[0018] Obtain the installation height of the on-vehicle camera, the slope difference between the plane where the vehicle is currently located and the plane where the line parking space is located, the virtual corner point information with errors, and the corner point information of the real line parking space. Through MATLAB correlation analysis, obtain the quadratic relational expression between the virtual corner point information with errors and the corner point information of the real line parking space, and this quadratic relational expression is the correction relational expression.

[0019] A further technical solution is that the correction relational expression is:

[0020] Ahx 2 +Bhx + C = w2;

[0021] Where x = w1, w1 is the width of the virtual parking space corner point with errors obtained by conventional projection based on the corner point information, w2 is the width of the corner point of the real line parking space, h is the installation height of the on-vehicle camera, and the parameters A and B are parameters related to the slope difference a, obtained by data fitting.

[0022] A further technical solution is that determining the virtual parking space lines projected onto the plane where the line parking space is located based on the corrected virtual corner point information includes:

[0023] Based on the corrected virtual corner point information, the smooth corner point positions are obtained through Kalman filtering processing;

[0024] Using a geometric correction algorithm, rectangular correction is performed on the smooth corner point positions, specifically: connecting the corner points to form a quadrilateral, dividing the quadrilateral into two triangles, correcting the two triangles into right triangles, and updating the corner point positions according to the correction relationship to complete the rectangular correction of the corner point positions;

[0025] Projection is performed according to the corrected corner point positions to obtain the virtual parking space lines on the plane where the line parking spaces are located.

[0026] In a second aspect, the present invention provides a ramp line parking space position determination system based on visual correction.

[0027] A ramp line parking space position determination system based on visual correction includes:

[0028] A corner point information acquisition module, which is used to obtain the original image of the line parking space by using an in-vehicle camera, identify the line parking space in the image through an image recognition algorithm, and determine the corner point information of the line parking space in the original image; wherein, the corner point information includes the position information of the four corner points of the line parking space;

[0029] A parking space line projection module, which is used to obtain the slope difference between the plane where the current vehicle is located and the plane where the line parking space is located. If the slope difference is equal to 0, the virtual parking space lines are directly projected according to the corner point information of the line parking space; if the slope difference is not equal to 0, the corner point information of the line parking space is corrected, and then the virtual parking space lines are projected according to the corrected corner point information;

[0030] A line parking space position determination module, which is used to determine the real position of the line parking space on the plane where it is located according to the virtual parking space lines.

[0031] In a third aspect, the present invention further provides an electronic device, including a memory, a processor, and computer instructions stored on the memory and running on the processor. When the computer instructions are run by the processor, the steps of the method described in the first aspect are completed.

[0032] In a fourth aspect, the present invention further provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by the processor, the steps of the method described in the first aspect are completed.

[0033] In a fifth aspect, the present invention further provides a vehicle. When the vehicle performs automatic parking, the position of the line parking space is determined by using a ramp line parking space position determination method based on visual correction as described in the first aspect, or the vehicle includes a ramp line parking space position determination system based on visual correction as described in the second aspect.

[0034] The above one or more technical solutions have the following beneficial effects:

[0035] 1. The present invention provides a method, a system and a vehicle for determining the position of a ramp line parking space based on vision correction. When the vehicle and the parking space are not on the same ground plane or horizontal plane, a vision recognition algorithm is used to recognize the ramp line parking space, and combined with the slope difference, the recognized ramp line parking space is corrected for errors, so as to accurately determine the position on the plane where the ramp line parking space is located, which can greatly improve the parking success rate in the slope scenario and reduce the automatic parking time.

[0036] 2. In the method for determining the position of a ramp line parking space based on vision correction proposed by the present invention, according to the slope difference between the actual vehicle and the plane where the line parking space is located, it is judged whether it is necessary to correct the corner point information of the line parking space in the vision recognition image. If there is a slope difference, the corner point information is corrected by combining a pre-constructed correction relationship, and an accurate virtual parking line is generated by projecting according to the corrected corner point information, so as to determine the position of the line parking space on the plane where it is located, ensuring the accurate recognition and determination of the position of the ramp line parking space, laying a foundation for subsequent vehicle automatic parking, and improving the automatic parking success rate. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The accompanying drawings forming a part of this invention are used to provide a further understanding of the invention. The schematic embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0038] Figure 1 is a flowchart of the method for determining the position of a ramp line parking space based on vision correction according to an embodiment of the present invention;

[0039] Figure 2 is a schematic diagram of the projection of the corner point information of the line parking space when the slope difference is 0 in an embodiment of the present invention;

[0040] Figure 3 is a schematic diagram of the projection of the corner point information of the line parking space when the slope difference is not 0 in an embodiment of the present invention;

[0041] Figure 4 is a schematic flowchart of correcting the corner point information of the line parking space in an embodiment of the present invention;

[0042] Figure 5 is a schematic diagram of rectangle correction in an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0043] It should be noted that the following detailed description is exemplary only for describing specific embodiments, aiming to provide further illustration of the present invention and not intended to limit the exemplary embodiments of the present invention. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. In addition, it should also be understood that when the terms "comprising" and / or "including" are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0044] Embodiment 1

[0045] This embodiment provides a method for determining the position of a ramp line parking space based on vision correction, as Figure 1 shown, including the following steps:

[0046] Step S1: Use an in-vehicle camera to obtain the original image of the line parking space, identify the line parking space in the image through an image recognition algorithm, and determine the corner point information of the line parking space in the original image; wherein, the corner point information includes the position information of the four corner points of the line parking space.

[0047] Step S2: Obtain the slope difference between the plane where the current vehicle is located and the plane where the line parking space is located. If the slope difference is equal to 0, directly project the corner point information of the line parking space as virtual parking space lines; if the slope difference is not equal to 0, correct the corner point information of the line parking space, and then project the corrected corner point information as virtual parking space lines.

[0048] Step S3: Determine the true position of the line parking space on the plane where it is located according to the virtual parking space lines.

[0049] The method for determining the position of a ramp line parking space based on vision correction proposed in this embodiment is introduced in more detail through the following content.

[0050] In step S1, use the in-vehicle cameras mounted around the vehicle to obtain the original image of the line parking space, and use an image recognition algorithm (such as convolutional neural network CNN, object detection algorithm YOLO, etc.) to identify the line parking space in the image, and determine the corner point information of the line parking space in the original image. This corner point information includes the position information of the four corner points of the line parking space.

[0051] As an implementation method, build an object detection model based on a convolutional neural network, construct a training data set with the images labeled with the line parking space and its respective corner points, use this training data set to train the built object detection model, and then the trained network model can be used to identify the line parking space in the image, output the pixel-level line parking space and the corner points of the line parking space in the original image. Among them, the line parking space and the corner points of the line parking space are both obtained by fitting pixel points. Further, according to the identified line parking space and the corner points of the line parking space, determine the corner point information of the line parking space in the original image.

[0052] After obtaining the corner point information of the line parking space in the original image, the corner point information is preprocessed. Specifically, the pixel-level line parking space and its corner points in the original image are identified using an image recognition algorithm. Considering that there may be certain errors in the image recognition pixel points, this embodiment first performs a preprocessing operation on the recognition result, including: first, for all the identified pixel points, the distance of each pixel point from the pixel-level line parking space corner point output by the first fitting of the above target detection model is calculated, and the pixel points whose distance exceeds the set distance (the distance is set to 5cm in this embodiment) are eliminated; then, a secondary fitting is performed based on the eliminated pixel points to output accurate line parking space corner points, so as to avoid the adverse effects of recognition errors on subsequent data processing.

[0053] In step S2, the slope difference between the plane where the current vehicle is located and the plane where the linear parking space is located is obtained, and whether to correct the corner point information is determined according to the slope difference, including: if the slope difference is equal to 0, directly projecting the corner point information of the linear parking space into a virtual parking space line; if the slope difference is not equal to 0, correcting the corner point information of the linear parking space, and then projecting the corrected corner point information into a virtual parking space line.

[0054] Specifically, the vehicle obtains the slope of the current position of the vehicle through its own inertial sensor IMU, that is, obtains the slope of the plane where the vehicle is currently located; at the same time, the vehicle is also equipped with a three-dimensional laser scanning device, which scans the entire surface of the plane where the linear parking space is located, so as to obtain the accurate slope of the plane where the linear parking space is located; preferably, the vehicle can also be equipped with measuring instruments such as a slope meter or a theodolite, and use these instruments to measure and obtain the slope of the plane where the linear parking space is located. On this basis, the slope difference is calculated based on the two slopes obtained by measurement.

[0055] Furthermore, when the slope difference is equal to 0, the current vehicle and the line parking space are at the same horizontal plane / ground plane, such as Figure 2 As shown, since the installation position of the vehicle body camera is fixed, the line parking space corner point information in the recognized original image can be directly projected onto the horizontal plane / ground plane, and the actual position of the line parking space corner point can be obtained by calculating the geometric relationship in the projection process to generate a virtual parking space line.

[0056] Specifically, the geometric relationship in the above projection process is calculated as follows: first, based on the camera ranging principle, a mathematical model is constructed according to the camera focal length (f), the position of the line parking space corner point in the image, the actual position of the line parking space corner point, and the actual distance D between the camera and the line parking space; secondly, based on the mathematical model constructed above, according to the identified line parking space corner point position information, the actual position of the line parking space is calculated and determined, and a virtual parking space line is generated through projection.

[0057] Further, when the slope difference is not equal to 0, that is, the current vehicle and the line parking space are not on the same horizontal plane / ground plane. For example, Figure 3 As shown, if a projection is directly performed at this time, a projection error will occur, and the projected parking space width w2 is not equal to the actual width w1 of the line parking space. In fact, for standard parking space lines, the slope difference will cause a ghosting error of 15 - 25 cm. Therefore, in this embodiment, when the vehicle and the parking space are not on the same horizontal plane / ground plane and there is a slope difference between the plane where the vehicle is located and the plane where the parking space is located, the corner point information of the line parking space is corrected and then projected as a virtual parking space line to ensure that the virtual parking space line generated by the projection is the same as the actual parking space line.

[0058] Specifically, the above-mentioned process of correcting the corner point information of the line parking space and projecting it as a virtual parking space line based on the corrected information is as follows: Figure 4 As shown, based on the obtained preprocessed corner point information, first, the corner point information of the line parking space in the original image is projected onto the plane where the vehicle is currently located to obtain virtual corner point information with errors. It should be noted that during the above projection process, the plane where the original image is located is always perpendicular to the plane where the vehicle is currently located; second, according to the installation height h of the on-vehicle camera, the slope difference a between the plane where the vehicle is currently located and the plane where the line parking space is located, and the virtual corner point information with errors, substitute them into a preset correction relational expression to compensate for the position of the corner points and output the corrected virtual corner point information to eliminate the slope error through correction; finally, project according to the corrected virtual corner point information to obtain the virtual parking space line of the plane where the line parking space is located.

[0059] Among them, the determination of the above correction relational expression is pre-determined. Specifically, the correction relational expression is constructed by the method of correlation learning, and then the slope error is compensated based on this correlation relationship to eliminate the misalignment of parking space recognition and improve the accuracy of parking space recognition. The construction of the correction relational expression specifically includes:

[0060] First, obtain the installation height h of the on-vehicle camera, the slope difference a between the plane where the vehicle is currently located and the plane where the line parking space is located, the virtual corner point information with errors, and the corner point information of the real line parking space. In this embodiment, for the virtual corner point information with errors and the corner point information of the real line parking space, the calculation is performed taking the width w1 of the virtual parking space corner point with errors obtained by normal projection and the width w2 of the corner point of the real line parking space as an example.

[0061] Secondly, about 200 groups of the above-mentioned various parameter data are collected, and correlation analysis is carried out based on each group of calibrated data. Through MATLAB correlation analysis, a quadratic relationship is obtained between the virtual corner point information with errors and the corner point information of the real line parking space, that is, the quadratic relationship between the corner point width w2 of the real line parking space, the corner point width w1 of the virtual parking space with errors obtained by conventional projection based on the corner point information, the installation height h of the on-vehicle camera, and the slope difference a between the plane where the vehicle is currently located and the plane where the line parking space is located. This quadratic relationship is the calibration relationship and can be expressed as:

[0062] Ahx 2 +Bhx+C = w2;

[0063] where x = w1, w1 is the corner point width of the virtual parking space with errors obtained by conventional projection based on the corner point information, w2 is the corner point width of the real line parking space, h is the installation height of the on-vehicle camera, and the parameters A and B are parameters related to the slope difference a. These parameters A and B can be obtained by fitting a large amount of experimental data.

[0064] Preferably, considering that the parking space is in dynamic change relative to the vehicle body, therefore, during the process of projecting according to the calibrated virtual corner point information, first, the smooth corner point position is obtained through Kalman filtering processing, which can effectively remove baseline drift and ensure the accuracy of subsequent calculations; then, using the geometric calibration algorithm, the smooth corner point position is rectified into a rectangle. As Figure 5 shown, a quadrilateral is formed by connecting the corner points, the quadrilateral is cut into two triangles, and the two triangles are rectified into right triangles to update the corner point position, thereby completing the rectangular rectification of the corner point position; finally, projection is carried out according to the rectified corner point position to obtain the virtual parking space on the plane where the line parking space is located.

[0065] In step S3, according to the virtual parking space line identified and determined in the above steps, the real position of the line parking space on the plane where it is located is determined, and automatic parking is carried out based on the accurately identified line parking space, improving the accuracy and efficiency of automatic parking.

[0066] Embodiment 2

[0067] This embodiment provides a ramp line parking space position determination system based on vision correction, including:

[0068] A corner point information acquisition module, configured to use an on-vehicle camera to acquire the original image of the line parking space, identify the line parking space in the image through an image recognition algorithm, and determine the corner point information of the line parking space in the original image; wherein, the corner point information includes the position information of the four corner points of the line parking space.

[0069] The parking space line projection module is used to obtain the slope difference between the plane where the current vehicle is located and the plane where the line parking space is located. If the slope difference is equal to 0, the virtual parking space line is directly projected according to the corner point information of the line parking space; if the slope difference is not equal to 0, the corner point information of the line parking space is corrected, and then the virtual parking space line is projected according to the corrected corner point information.

[0070] The line parking space position determination module is used to determine the real position of the line parking space on the plane where it is located according to the virtual parking space line.

[0071] Embodiment III

[0072] This embodiment provides an electronic device, including a memory, a processor, and computer instructions stored on the memory and running on the processor. When the computer instructions are run by the processor, the steps in the above-mentioned method for determining the position of a ramp line parking space based on vision correction are completed.

[0073] Embodiment IV

[0074] This embodiment also provides a computer-readable storage medium for storing computer instructions. When the computer instructions are executed by the processor, the steps in the above-mentioned method for determining the position of a ramp line parking space based on vision correction are completed.

[0075] Embodiment V

[0076] This embodiment provides a vehicle. When the vehicle performs automatic parking, the method for determining the position of the line parking space based on vision correction as described above is used to determine the position of the line parking space, or the vehicle includes a system for determining the position of the ramp line parking space based on vision correction as described above.

[0077] The steps involved in Embodiments II to V above correspond to those in Method Embodiment I. For specific implementation manners, reference may be made to the relevant description part of Embodiment I. The term "computer-readable storage medium" should be understood to include a single medium or multiple media containing one or more instruction sets; it should also be understood to include any medium that can store, encode, or carry an instruction set for execution by a processor and cause the processor to execute any method in the present invention.

[0078] Those skilled in the art should understand that the above-mentioned modules or steps of the present invention can be implemented by a general-purpose computer device. Optionally, they can be implemented by program codes executable by a computing device, so that they can be stored in a storage device and executed by the computing device, or they can be separately fabricated into individual integrated circuit modules, or multiple modules or steps among them can be fabricated into a single integrated circuit module for implementation. The present invention is not limited to any specific combination of hardware and software.

[0079] The above are only the preferred embodiments of the present invention. Although the specific implementation manners of the present invention are described in conjunction with the accompanying drawings, it is not a limitation on the protection scope of the present invention. Those skilled in the art should understand that various modifications or deformations that can be made without creative efforts on the basis of the technical solution of the present invention are still within the protection scope of the present invention.

Claims

1. A method for determining the parking position of a ramp line based on visual correction, characterized in that: include: The original image of the linear parking space is obtained by using the vehicle-mounted camera, the linear parking space in the image is identified by using the image recognition algorithm, and the corner point information of the linear parking space in the original image is determined; wherein the corner point information includes the position information of the four corner points of the linear parking space; Get the slope difference between the plane where the current vehicle is located and the plane where the line parking space is located. If the slope difference is equal to 0, directly project the corner point information of the line parking space into a virtual parking space line; if the slope difference is not equal to 0, correct the corner point information of the line parking space, and then project the corrected corner point information into a virtual parking space line; According to the virtual parking space line, determine the actual position of the line parking space on the plane.

2. A method for determining the parking position of a ramp line based on visual correction as claimed in claim 1, characterized in that: After obtaining the corner point information of the line parking space in the original image, the corner point information is preprocessed, including: Abnormal point removal is: for all identified pixels, the distance between each pixel and the fitted pixel-level line parking space corner point is calculated, and the pixels whose distance exceeds the set distance are removed; Refitting is to perform secondary fitting based on the removed pixels and output accurate line parking space corner point information.

3. The method for determining the parking position of a ramp line based on visual correction according to claim 1, characterized in that: Correct the corner point information of the parking space line, and then project the corrected corner point information into a virtual parking space line, including: Project the corner point information of the line parking space in the original image to the plane where the current vehicle is located, and obtain virtual corner point information with errors; wherein the plane where the original image is located is always perpendicular to the plane where the current vehicle is located; According to the installation height of the vehicle-mounted camera, the slope difference between the plane where the current vehicle is located and the plane where the linear parking space is located, and the virtual corner point information with errors, they are substituted into the preset correction relationship to output the corrected virtual corner point information; According to the corrected virtual corner point information, a virtual parking space line projected onto the plane where the linear parking space is located is determined.

4. The method for determining the parking position of a ramp line based on visual correction according to claim 3, characterized in that: The determination of the correction relationship includes: Get the installation height of the vehicle camera, the slope difference between the plane where the vehicle is currently located and the plane where the linear parking space is located, the virtual corner point information with errors, and the corner point information of the real linear parking space. Use MATLAB correlation analysis to get the quadratic relationship between the virtual corner point information with errors and the corner point information of the real linear parking space. This quadratic relationship is the correction relationship.

5. The method for determining the parking position of a ramp line based on visual correction according to claim 4, characterized in that: The correction relationship is: Ahx 2 +Bhx+C=w2; Among them, x=w1, w1 is the virtual parking space corner point width with error obtained by conventional projection based on corner point information, w2 is the real line parking space corner point width, h is the installation height of the vehicle-mounted camera, and parameters A and B are parameters related to the slope difference a, which are obtained through data fitting.

6. The method for determining the parking position of a ramp line based on visual correction as claimed in claim 3, characterized in that: According to the corrected virtual corner point information, determine the virtual parking space line projected onto the plane where the linear parking space is located, including: Based on the corrected virtual corner point information, the smoothed corner point position is obtained through Kalman filtering; The geometric correction algorithm is used to perform rectangular correction on the smooth corner point positions, which is as follows: a quadrilateral is formed according to the lines connecting the corner point positions, the quadrilateral is cut and divided into two triangles, the two triangles are corrected into right triangles, the corner point positions are updated according to the correction relationship, and the rectangular correction of the corner point positions is completed; Projection is performed according to the corrected corner point positions to obtain a virtual parking space line on the plane where the linear parking space is located.

7. A system for determining the parking position of a ramp line based on visual correction, characterized in that: include: The corner point information acquisition module is used to acquire the original image of the linear parking space by using the vehicle-mounted camera, identify the linear parking space in the image by using the image recognition algorithm, and determine the corner point information of the linear parking space in the original image; wherein the corner point information includes the position information of the four corner points of the linear parking space; The parking space line projection module is used to obtain the slope difference between the plane where the current vehicle is located and the plane where the line parking space is located. If the slope difference is equal to 0, it is directly projected as a virtual parking space line according to the corner point information of the line parking space; if the slope difference is not equal to 0, the corner point information of the line parking space is corrected, and then projected as a virtual parking space line according to the corrected corner point information; The line parking space position determination module is used to determine the real position of the line parking space on the plane according to the virtual parking space line.

8. An electronic device, characterized in that: The method comprises a memory and a processor and computer instructions stored in the memory and executed on the processor. When the computer instructions are executed by the processor, the steps of a method for determining a parking space position on a ramp line based on visual correction as described in any one of claims 1 to 6 are completed.

9. A computer-readable storage medium, characterized in that: Used to store computer instructions, which, when executed by a processor, complete the steps of a method for determining a ramp line parking space position based on visual correction as described in any one of claims 1-6.

10. A vehicle, characterized in that: When the vehicle is performing automatic parking, the line parking position is determined by using a ramp line parking position determination method based on visual correction as described in any one of claims 1-6, or the vehicle includes a ramp line parking position determination system based on visual correction as described in claim 7.

Images