Virtual lane line generation method and device, and camera

Abstract

The invention relates to a virtual lane line generation method. The method comprises the following steps: checking whether a current scene meets a generation condition of an intersection virtual lane line or not; when the generation condition is satisfied, generating a virtual lane line based on the left and right lane lines of the current lane; and connecting the generated virtual lane line with the opposite lane line of the intersection. The invention further relates to a computer storage medium, a computer program product, virtual lane line generation equipment and a camera.

Description

Technical Field

[0001] The present application relates to the field of virtual lane line generation, and more specifically, to a virtual lane line generation method and device, a computer storage medium, a computer program product, and a camera. Background Art

[0002] Lane recognition is an important part of the driver assistance system, because the steering wheel function control requires accurate lane information to achieve vehicle speed and posture control. For scenes with clear lane lines on the road, we can use deep learning algorithms or traditional feature-based algorithms to recognize lane lines, but for intersections without lane lines, the assisted driving function cannot be maintained because the lane lines cannot be recognized.

[0003] In order to ensure that the assisted driving function does not exit, the existing solution for intersection scenarios is generally based on the car-following theory. However, the existing solution obviously cannot be triggered or performed when there is no car in front. At the same time, the safety and robustness of the existing solution are also difficult to guarantee. Summary of the invention

[0004] To address the problem of missing lane lines in intersection scenarios, one or more embodiments of the present application provide a solution for generating virtual lane lines at intersections based on image vision.

[0005] According to one aspect of the present application, a method for generating a virtual lane line is provided, the method comprising: checking whether a current scene satisfies a generation condition for a virtual lane line at an intersection; when the generation condition is met, generating a virtual lane line based on left and right lane lines of the current lane; and connecting the generated virtual lane line with an opposite lane line at the intersection.

[0006] As a supplement or replacement for the above solution, the above method may also include: after an exit condition is met, exiting the virtual lane line generation method.

[0007] As a supplement or replacement for the above-mentioned scheme, in the above-mentioned method, checking whether the current scene meets the generation conditions of the virtual lane lines at the intersection includes: obtaining the real-time position and posture information of the vehicle to determine that the vehicle is going straight; determining that the left and right lane lines of the lane where the vehicle is located exist and the curvature of the lane is within a first preset threshold range; identifying the zebra crossing at the intersection, and setting the zebra crossing sign to valid when the zebra crossing is identified, otherwise setting the zebra crossing sign to invalid; and when the zebra crossing sign is set to valid, if the preview distance of the left and right lane lines is less than the farthest distance of the zebra crossing, or if the preview distance of the left and right lane lines is greater than or equal to the farthest distance of the zebra crossing and the left and right lane lines have not been updated in a predetermined number of consecutive image frames, it is determined that the current scene meets the generation conditions of the virtual lane lines at the intersection.

[0008] As a supplement or replacement for the above-mentioned scheme, in the above-mentioned method, checking whether the current scene meets the generation conditions of the virtual lane lines at the intersection also includes: when the zebra crossing sign is set to invalid, if the zebra crossing sign is valid in the previous image frame, setting the intersection sign to valid; and when the zebra crossing sign is set to invalid, if the zebra crossing sign is valid in the previous image frame and the left and right lane lines have not been updated in a consecutive predetermined number of image frames, or if the zebra crossing sign is invalid in the previous image frame but the intersection sign is set to valid and the left and right lane lines have not been updated in a consecutive predetermined number of image frames, determining that the current scene meets the generation conditions of the virtual lane lines at the intersection.

[0009] As a supplement or replacement for the above-mentioned scheme, in the above-mentioned method, when the generation condition is met, generating a virtual lane line based on the left and right lane lines of the current lane includes: checking whether the left and right lane lines of the current lane are identified and whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; when the left and right lane lines of the current lane are identified and the intersection function flag is valid, checking whether the preview distance of the left and right lane lines is greater than a second preset threshold; when it is determined that the left and right lane lines have not been updated in a consecutive predetermined number of image frames, generating the virtual lane line based on the historical lane line; and updating the attributes of the virtual lane line.

[0010] As a supplement or replacement for the above-mentioned scheme, in the above-mentioned method, connecting the generated virtual lane line with the opposite lane line of the intersection includes: checking whether the left and right lane lines of the current lane are identified, and whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; when the opposite lane line of the intersection is identified, using the opposite lane line of the intersection to update the virtual lane line so as to obtain a new virtual lane line; and when the lateral deviation between the new virtual lane line and the virtual lane line before the update is within a third preset threshold, making the left and right lane lines in the new virtual lane line parallel to each other.

[0011] As a supplement or replacement for the above-mentioned scheme, in the above-mentioned method, after the exit condition is met, exiting the virtual lane line generation method includes: checking whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; and after the intersection function flag is set to be valid, when the driving distance and driving time of the vehicle exceed a fourth preset threshold, determining that the exit condition is met and exiting the virtual lane line generation method.

[0012] As a supplement or replacement for the above-mentioned scheme, in the above-mentioned method, after the exit condition is met, exiting the virtual lane line generation method also includes: when the driving distance and driving time of the vehicle do not exceed the fourth preset threshold after the intersection function flag is set to valid, but the curb and the lane line of the adjacent lane can be stably identified, checking the zebra crossing at the intersection and the opposite lane line at the intersection; and when the zebra crossing is identified, or when the opposite lane line at the intersection is identified and the virtual lane line is successfully updated using the opposite lane line at the intersection, determining that the exit condition is met and exiting the virtual lane line generation method.

[0013] According to another aspect of the present application, a computer storage medium is provided, wherein the medium includes instructions, and the instructions execute the above method when executed.

[0014] According to another aspect of the present application, a computer program product is provided, comprising a computer program, and when the computer program is executed by a processor, the method described above is implemented.

[0015] According to another aspect of the present application, a virtual lane line generation device is provided, the device comprising: a checking device for checking whether the current scene satisfies the generation conditions of the virtual lane line at the intersection; a generating device for generating the virtual lane line based on the left and right lane lines of the current lane when the generation conditions are met; and a connecting device for connecting the generated virtual lane line with the opposite lane line of the intersection.

[0016] As a supplement or replacement for the above solution, the above device also includes: an exit device, which is used to close the virtual lane line generating device after the exit condition is met.

[0017] According to another aspect of the present application, a camera is provided, comprising: an acquisition device for acquiring data related to a current scene; and a virtual lane line generation device as described above, for generating a virtual lane line based on the data related to the current scene.

[0018] The virtual lane line generation scheme of the embodiment of the present application checks whether the current scene meets the generation conditions of the virtual lane line at the intersection, and when the generation conditions are met, generates the virtual lane line based on the left and right lane lines of the current lane and connects the generated virtual lane line with the opposite lane line of the intersection, thereby providing a virtual lane line generation scheme for the intersection based on image vision without the need for additional hardware. In addition, by connecting the generated virtual lane line with the opposite lane line of the intersection, the lane line accuracy can be continuously adjusted in real time using the opposite lane line of the intersection. In addition, in one embodiment, the virtual lane line is generated based on the historical lane line, which can effectively ensure the reliability of the generated virtual lane line.

[0019] The virtual lane line generation solution of the embodiment of the present application can be applied to various intersections, including ordinary crossroads, three-way non-right-angle intersections, etc. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] The above and other objects and advantages of the present application will become more fully apparent from the following detailed description in conjunction with the accompanying drawings, wherein the same or similar elements are represented by the same reference numerals.

[0021] Figure 1 A schematic diagram of a process of generating a virtual lane line according to an embodiment of the present application is shown;

[0022] Figure 2 A schematic diagram of the structure of a virtual lane line generating device according to an embodiment of the present application is shown;

[0023] Figure 3 A schematic diagram of a process for checking whether a current scene satisfies a generation condition of a virtual lane line at an intersection according to an embodiment of the present application is shown;

[0024] Figure 4 A schematic diagram of a process of generating a virtual lane line based on the left and right lane lines of the current lane when the generation condition is met according to an embodiment of the present application is shown;

[0025] Figure 5 A schematic diagram of a process of connecting a generated virtual lane line with an opposite lane line at an intersection according to an embodiment of the present application is shown;

[0026] Figure 6 A schematic diagram showing parallelism detection of new virtual lane lines according to an embodiment of the present application is shown;

[0027] Figure 7 A schematic diagram showing a flow chart of exiting the virtual lane line generation method after an exit condition is met according to an embodiment of the present application; and

[0028] Figure 8 A schematic structural diagram of a camera according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0029] Hereinafter, a virtual lane line generation scheme according to various exemplary embodiments of the present application will be described in detail with reference to the accompanying drawings.

[0030] Figure 1 FIG. 1 is a flow chart of a method 1000 for generating a virtual lane line according to an embodiment of the present application. Figure 1 As shown, the virtual lane line generation method 1000 includes:

[0031] In step S110, it is checked whether the current scene meets the generation conditions of the virtual lane line at the intersection;

[0032] In step S120, when the generation condition is met, a virtual lane line is generated based on the left and right lane lines of the current lane; and

[0033] In step S130, the generated virtual lane line is connected to the opposite lane line of the intersection.

[0034] In the context of this application, the term "virtual lane line" is relative to "real lane line" and is applicable to intersection scenarios without lane lines, so it is also called "intersection virtual lane line". In one embodiment, virtual lane lines are generated in intersection scenarios to maintain the assisted driving function (avoid exiting the assisted driving function).

[0035] Continue to refer Figure 1 In an optional embodiment, the above-mentioned virtual lane line generation method 1000 may further include step S140, wherein the virtual lane line generation method is exited after the exit condition is met.

[0036] In step S110, it is checked whether the current scene meets the generation conditions of the virtual lane line at the intersection. Here, the relevant information of the "current scene" is obtained through various sensors (such as cameras) based on image vision. In one embodiment, the camera installed on the vehicle is used to obtain road information in real time, including but not limited to: lane line information, zebra crossing information, etc.

[0037] The step S110 may further include the following sub-steps: obtaining the real-time position and posture information of the vehicle (for example, obtaining in real time through SLAM (Simultaneous Localization and Mapping) related algorithms) to determine that the vehicle is going straight; determining that the left and right lane lines of the lane where the vehicle is located exist (for example, whether they are recognized by the current algorithm), and the curvature of the lane is within the first preset threshold range; identifying the zebra crossing at the intersection, and setting the zebra crossing flag to valid when the zebra crossing is recognized, otherwise setting the zebra crossing flag to invalid; and when the zebra crossing flag is set to valid, if the preview distance of the left and right lane lines (or the length of the left and right lane lines) is less than the farthest distance of the zebra crossing, or if the preview distance of the left and right lane lines is greater than or equal to the farthest distance of the zebra crossing and the left and right lane lines are not updated in a predetermined number of consecutive image frames (for example, more than two image frames), it is determined that the current scene meets the generation conditions of the virtual lane lines at the intersection. Here, the "zebra crossing flag" indicates whether the zebra crossing is recognized. If it is recognized, the "zebra crossing flag" is set to valid, otherwise it is set to invalid.

[0038] In one embodiment, step S110 may also include the following sub-steps: when the zebra crossing sign is set to invalid, if the zebra crossing sign is valid in the previous image frame, the intersection sign is set to valid; and when the zebra crossing sign is set to invalid, if the zebra crossing sign is valid in the previous image frame and the left and right lane lines are not updated in a continuous predetermined number of image frames (e.g., more than two image frames), or if the zebra crossing sign is invalid in the previous image frame but the intersection sign is set to valid and the left and right lane lines are not updated in a continuous predetermined number of image frames (e.g., more than two image frames), it is determined that the current scene meets the generation conditions of the intersection virtual lane line. Here, the "intersection sign" indicates whether the current scene is identified as an intersection. If the current scene is identified as an intersection, the "intersection sign" is set to valid, otherwise it is set to invalid.

[0039] Figure 3 FIG. 1 is a flow chart showing a process of checking whether the current scene satisfies the generation conditions of the virtual lane line at the intersection according to an embodiment of the present application. Figure 3 As shown, in step S310, the posture of the vehicle is obtained to ensure that the vehicle is going straight. For example, the real-time position and posture information of the vehicle is obtained using the SLAM-related algorithm. It is worth noting that the method of this embodiment is only effective when the vehicle is going straight, that is, if it is determined that the vehicle body has a turning action, the subsequent steps will not be executed. In step S320, the attributes of the left and right lane lines of the vehicle are obtained to ensure that they are valid and straight. For example, it can be determined whether the current algorithm recognizes the left and right lane lines. Only when there are left and right lane lines and the curvature is within the set threshold range, the subsequent steps will continue to be executed, otherwise it is determined that the current scene does not meet the generation conditions of the virtual lane lines (the sign of the intersection function will not be set). In step S330, based on (but not limited to) the deep learning algorithm, it is determined whether the intersection zebra crossing is identified in the image of the current frame; if so, step S340 is executed to set the zebra crossing flag to be valid; otherwise, step S350 is executed to set the zebra crossing flag to be invalid. After executing step S340 to set the zebra crossing flag to be valid, step S360 is executed. In step S360, determine whether the preview distance of the left and right lane lines is less than the farthest distance of the zebra crossing; if so, execute step S380 to determine that the current scene meets the generation conditions of the virtual lane lines at the intersection (the "intersection function (on) flag" is set to be valid); otherwise (i.e., the preview distance of the left and right lane lines is greater than or equal to the farthest distance of the zebra crossing), execute step S375 to determine whether the left and right lane lines have not been updated for more than two image frames, and if so, continue to execute step S380; otherwise, determine that the current scene does not meet the generation conditions of the virtual lane lines at the intersection (i.e., jump out of the Figure 3 Execution flow shown).

[0040] After executing step S350 to set the zebra crossing flag to invalid, execute step S365. In step S365, determine whether the zebra crossing flag is valid in the previous image frame; if valid, execute step S375, and if invalid, execute step S370. In step S370, determine whether the intersection flag is turned on (valid), if valid, execute step S375, otherwise determine that the current scene does not meet the conditions for generating the intersection virtual lane line (i.e., jump out of the scene). Figure 3 In step S375, as described above, it is determined whether the left and right lane lines have not been updated for more than two image frames. If so, the process continues to step S380; otherwise, it is determined that the current scene does not meet the generation conditions of the virtual lane lines at the intersection (i.e., the process is jumped out of step S380). Figure 3 In step S380, (since all checks have been passed) it is determined that the current scene meets the conditions for generating the virtual lane line at the intersection (the "intersection function (on) flag" is set to be valid).

[0041] In step S120, when the generation condition is met, a virtual lane line is generated based on the left and right lane lines of the current lane. That is, this step is responsible for creating a virtual lane line at the intersection. In one embodiment, step S120 includes: checking whether the left and right lane lines of the current lane are identified, and whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition of the virtual lane line at the intersection is met; when the left and right lane lines of the current lane are identified, and the intersection function flag is valid, checking whether the preview distance of the left and right lane lines is greater than a second preset threshold; when it is determined that the left and right lane lines have not been updated in a predetermined number of consecutive image frames (for example, more than two image frames), generating the virtual lane line based on the historical lane line; and updating the attributes of the virtual lane line.

[0042] Figure 4 FIG. 1 is a flow chart showing a process of generating virtual lane lines based on the left and right lane lines of the current lane when the generation conditions are met according to an embodiment of the present application. Figure 4 As shown, in step S410, it is checked whether the left and right lane lines are recognized in the current image frame, and whether the intersection function flag is valid. If both are valid, then continue to step S420, otherwise exit. Figure 4The process of generating virtual lane lines is shown in FIG. 4. In step S420, check whether the length of the left and right lane lines in the current intersection scene (i.e., the preview distance) meets the requirements (e.g., whether it is greater than the second preset threshold). If so, execute step S430; otherwise, execute step S425. In step S425, set the lane line length to the default value. In step S430, further determine whether the real lane line is identified. If the real lane line is not identified for two consecutive frames (i.e., the left and right lane lines are not updated for more than two image frames), proceed to step S440, otherwise exit. Figure 4 The process of generating virtual lane lines is shown in FIG. 4. In step S440, considering that the current scene is an intersection, since the lane lines are missing, virtual lane lines are created only based on the historical lane line attributes, so as to keep the assisted driving function from exiting. Finally, in step S450, the lane line attributes are updated, for example, the lane line shape is set to a solid line for safety reasons to prevent the vehicle from changing lanes.

[0043] exist Figure 4 In the illustrated embodiment, when the length of the left and right lane lines (i.e., the preview distance) in the current intersection scene is less than the second preset threshold, the lane line length is set to the default value, that is, the lane line needs to be lengthened. This is because, for example, the steering wheel control module (lateral function group) will use the lane line for control. If the lane line is too short, the steering wheel control module will consider the lane line to be less credible and will not use the lane line information for control.

[0044] In step S130, the generated virtual lane line is connected to the opposite lane line of the intersection. Here, the term "opposite lane line of the intersection" refers to the lane line on the lane opposite to the lane where the vehicle is located and located at the opposite end (the other end) of the intersection. This step is responsible for connecting the virtual lane line with the opposite lane line of the intersection to ensure the accuracy and reliability of the detection. In one embodiment, step S130 includes: checking whether the left and right lane lines of the current lane are identified, and whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; when the opposite lane line of the intersection is identified, the virtual lane line is updated using the opposite lane line of the intersection to obtain a new virtual lane line; and when the lateral deviation of the new virtual lane line and the virtual lane line before the update is within a third preset threshold, the left and right lane lines in the new virtual lane line are made parallel to each other.

[0045] Figure 5 FIG. 1 is a flow chart showing a process of connecting a generated virtual lane line with an opposite lane line at an intersection according to an embodiment of the present application. Figure 5As shown, in step S510, it is checked whether the left and right lane lines in the current image frame are valid (i.e., whether the left and right lane lines are recognized), and whether the intersection function flag is valid; if satisfied, continue to step S520, otherwise exit as shown in FIG. Figure 5 In step S520, the oncoming lane line of the intersection is identified based on (but not limited to) a deep learning algorithm and it is determined whether the identification is successful; if the oncoming lane line of the intersection is successfully identified, then step S530 is continued; otherwise, the process is exited. Figure 5 The process shown. In step S530, based on (but not limited to) filtering algorithms such as Kalman filtering and particle filtering, the curve equation parameters of the virtual lane line are updated using the real lane line identified in step S520. Next, in step S540, check whether the newly generated virtual lane line and the road position generated by the new virtual lane line meet the threshold requirements relative to the driving trajectory of the vehicle; if not, it is considered that the virtual lane line cannot be connected (matched) to the detected opposite lane line, and step S555 is executed, otherwise step S550 is executed. In step S550, when any angle at the preview distance of the new virtual lane line is greater than a preset threshold, the left and right lane lines in the new virtual lane line are made parallel to each other. In step S555, the valid bit of the lane line of the vehicle (i.e., the virtual lane line) is reset to make the virtual lane line invalid and deleted.

[0046] Figure 6 FIG. 2 shows a schematic diagram of performing parallelism detection on new virtual lane lines according to an embodiment of the present application. Figure 6 As shown, the vehicle 610 is traveling in a direction X on the lane, and the angles between the driving direction X and the left lane line 620 and the right lane line 625 at a certain point in the preview distance are θ1 and θ2 respectively. In one embodiment, the parallelism detection can be performed by checking the angle of a point at the end of the new connection line (i.e., the new virtual lane line) relative to the vehicle's forward direction, and obtaining the angles of the left and right lines respectively, thereby obtaining the angle difference information Δθ=abs(θ1-θ2). If the angle difference Δθ is greater than a set threshold (e.g., 8 degrees), the line that is most parallel to the vehicle's forward direction is used as a reference, and the other line is parallelized to prevent the virtual line from being randomly connected to the real line.

[0047] In step S140, after the exit condition is met, the virtual lane line generation method 1000 is exited. In one embodiment, step S140 includes: checking whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; and after the intersection function flag is set to be valid, when the driving distance and driving time of the vehicle exceed the fourth preset threshold, determining that the exit condition is met and exiting the virtual lane line generation method. In one embodiment, step S140 may also include: after the intersection function flag is set to be valid, when the driving distance and driving time of the vehicle do not exceed the fourth preset threshold, but the roadside and the lane line of the adjacent lane can be stably identified, checking the intersection zebra crossing and the intersection opposite lane line; and when the zebra crossing is identified, or when the intersection opposite lane line is identified and the virtual lane line is successfully updated using the intersection opposite lane line, determining that the exit condition is met and exiting the virtual lane line generation method.

[0048] Figure 7 FIG. 1 is a flow chart showing a process of exiting the virtual lane line generation method after an exit condition is met according to an embodiment of the present application. Figure 7 As shown, in step S710, whether the intersection function flag is valid, if so, proceed to step S720, otherwise exit as shown. Figure 7 In step S720, check whether the vehicle's driving distance and time after the function is turned on are within the threshold range. If so (i.e., not exceeding the threshold), proceed to step S730, otherwise proceed to step S750. In step S730, identify the roadside and lane lines of adjacent lanes based on (but not limited to) a deep learning algorithm, and determine whether stable identification can be achieved; if so, proceed to step S740, otherwise exit. Figure 7 In step S740, based on (but not limited to) the deep learning algorithm, the zebra crossing and the oncoming lane line of the intersection are identified, and it is determined whether the zebra crossing can be successfully identified or whether the oncoming lane line of the intersection can be identified and the virtual lane line is successfully updated using the oncoming lane line of the intersection; if so, the process proceeds to step S750, otherwise, the process exits as shown. Figure 7 In step S750, the intersection function flag is set to false (or invalid), that is, it is determined that the exit condition is met and the virtual lane line generation method is exited.

[0049] In one or more embodiments, the above steps S110, S120, S130 and S140 are executed when each frame of image arrives, and are repeated in a cycle.

[0050] The virtual lane line generation method of the above embodiment can solve the problem of missing lane lines in intersection scenes, and provide a solution for generating virtual lane lines at intersections based on image vision.

[0051] In addition, it is easy for a person skilled in the art to understand that the virtual lane line generation method 1000 provided in the above one or more embodiments of the present application can be implemented by a computer program. For example, the computer program is included in a computer program product, and when the computer program is executed by a processor, the virtual lane line generation method 1000 of one or more embodiments of the present application is implemented. For another example, when a computer storage medium (such as a USB flash drive) storing the computer program is connected to a computer, the virtual lane line generation method 1000 of one or more embodiments of the present application can be executed by running the computer program.

[0052] refer to Figure 2 , Figure 2 The structure diagram of a virtual lane line generating device 2000 according to an embodiment of the present application is shown. The virtual lane line generating device 2000 comprises: an inspection device 210, a generating device 220 and a connecting device 230. The inspection device 210 is used to check whether the current scene meets the generation conditions of the virtual lane line at the intersection; the generating device 220 is used to generate the virtual lane line based on the left and right lane lines of the current lane when the generation conditions are met; and the connecting device 230 is used to connect the generated virtual lane line with the opposite lane line at the intersection.

[0053] In an optional embodiment, the above-mentioned device 2000 also includes: an exit device 240, which is used to close the virtual lane line generating device 2000 after the exit condition is met.

[0054] In one or more embodiments, the inspection device 210 is configured to: obtain the real-time position and posture information of the vehicle to determine that the vehicle is going straight; determine that the left and right lane lines of the lane where the vehicle is located exist and that the curvature of the lane is within a first preset threshold range; identify the zebra crossing at the intersection, and set the zebra crossing sign to valid when the zebra crossing is identified, otherwise set the zebra crossing sign to invalid; and when the zebra crossing sign is set to valid, if the preview distance of the left and right lane lines is less than the farthest distance of the zebra crossing, or if the preview distance of the left and right lane lines is greater than or equal to the farthest distance of the zebra crossing and the left and right lane lines have not been updated in a predetermined number of consecutive image frames, determine that the current scene meets the generation conditions of virtual lane lines at the intersection.

[0055] In one or more embodiments, the inspection device 210 is further configured to: when the zebra crossing sign is set to invalid, if the zebra crossing sign was valid in the previous image frame, set the intersection sign to valid; and when the zebra crossing sign is set to invalid, if the zebra crossing sign was valid in the previous image frame and the left and right lane lines have not been updated in a consecutive predetermined number of image frames, or if the zebra crossing sign was invalid in the previous image frame but the intersection sign is set to valid and the left and right lane lines have not been updated in a consecutive predetermined number of image frames, determine that the current scene meets the generation conditions of the virtual lane lines at the intersection.

[0056] In one or more embodiments, the generating device 220 is configured to: check whether the left and right lane lines of the current lane are identified and whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generating condition is met; when the left and right lane lines of the current lane are identified and the intersection function flag is valid, check whether the preview distance of the left and right lane lines is greater than a second preset threshold; when it is determined that the left and right lane lines have not been updated in a predetermined number of consecutive image frames, generate the virtual lane lines based on the historical lane lines; and update the attributes of the virtual lane lines.

[0057] In one or more embodiments, the connecting device 230 is configured to: check whether the left and right lane lines of the current lane are identified, and whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; when the opposite lane line of the intersection is identified, the opposite lane line of the intersection is used to update the virtual lane line to obtain a new virtual lane line; and when the lateral deviation between the new virtual lane line and the virtual lane line before the update is within a third preset threshold, the left and right lane lines in the new virtual lane line are made parallel to each other.

[0058] In one or more embodiments, the exit device 240 is configured to: check whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; and after the intersection function flag is set to be valid, when the driving distance and driving time of the vehicle exceed a fourth preset threshold, determine that the exit condition is met and turn off the virtual lane line generation device 2000.

[0059] In one or more embodiments, the exit device 240 is also configured to: check the zebra crossing at the intersection and the opposite lane line at the intersection when the driving distance and driving time of the vehicle do not exceed the fourth preset threshold after the intersection function flag is set to valid, but the curb and the lane lines of the adjacent lanes can be stably identified; and when the zebra crossing is identified, or when the opposite lane line at the intersection is identified and the virtual lane line is successfully updated using the opposite lane line at the intersection, determine that the exit condition is met and turn off the virtual lane line generating device 2000.

[0060] The above-mentioned virtual lane line generating device 2000 can be integrated into various types of multifunctional cameras, including but not limited to MPC3, etc. In one embodiment, Figure 8 As shown, the camera 8000 includes: a collection device 810 and a virtual lane line generation device 820 (the specific functional structure of the virtual lane line generation device can be as shown in FIG. Figure 2 As shown), wherein the acquisition device 810 is used to acquire data related to the current scene; and the virtual lane line generation device 820 is used to generate a virtual lane line based on the data related to the current scene.

[0061] In summary, the virtual lane line generation scheme of the embodiment of the present application checks whether the current scene meets the generation conditions of the virtual lane line at the intersection, and when the generation conditions are met, generates the virtual lane line based on the left and right lane lines of the current lane and connects the generated virtual lane line with the opposite lane line of the intersection, thereby providing an image-based vision-based virtual lane line generation scheme for the intersection without the need for additional hardware. In addition, by connecting the generated virtual lane line with the opposite lane line of the intersection, the lane line accuracy can be continuously adjusted in real time using the opposite lane line of the intersection. In addition, in one embodiment, the virtual lane line is generated based on the historical lane line, which can effectively ensure the reliability of the generated virtual lane line.

[0062] The above examples mainly illustrate the virtual lane line generation scheme of the embodiments of the present application. Although only some of the implementation methods of the present application are described, it should be understood by those skilled in the art that the present application can be implemented in many other forms without departing from its subject matter and scope. Therefore, the examples and implementation methods shown are regarded as illustrative rather than restrictive, and the present application may cover various modifications and substitutions without departing from the spirit and scope of the present application as defined in the claims.

Claims

1. A method for generating a virtual lane line, characterized in that: The method comprises: Check whether the current scene meets the conditions for generating virtual lane lines at intersections; When the generation condition is met, generating a virtual lane line based on the left and right lane lines of the current lane; and Connect the generated virtual lane line with the opposite lane line of the intersection.

2. The method of claim 1, further comprising: After the exit condition is met, the virtual lane line generation method is exited.

3. The method of claim 1, wherein: Check whether the current scene meets the conditions for generating virtual lane lines at intersections: Acquiring the real-time position and posture information of the vehicle to determine whether the vehicle is moving straight; Determining that left and right lane lines of a lane where the vehicle is located exist and that a curvature of the lane is within a first preset threshold range; Identify a zebra crossing at an intersection, and when the zebra crossing is identified, set the zebra crossing flag to be valid, otherwise set the zebra crossing flag to be invalid; and When the zebra crossing flag is set to valid, if the preview distance of the left and right lane lines is less than the farthest distance of the zebra crossing, or if the preview distance of the left and right lane lines is greater than or equal to the farthest distance of the zebra crossing and the left and right lane lines are not updated in a predetermined number of consecutive image frames, it is determined that the current scene meets the conditions for generating virtual lane lines at the intersection.

4. The method of claim 3, wherein: Checking whether the current scene meets the conditions for generating virtual lane lines at intersections also includes: When the zebra crossing sign is set to invalid, if the zebra crossing sign was valid in the previous image frame, setting the intersection sign to valid; and When the zebra crossing sign is set to invalid, if the zebra crossing sign was valid in the previous image frame and the left and right lane lines have not been updated in a consecutive predetermined number of image frames, or if the zebra crossing sign was invalid in the previous image frame but the intersection sign is set to valid and the left and right lane lines have not been updated in a consecutive predetermined number of image frames, it is determined that the current scene meets the conditions for generating virtual lane lines at the intersection.

5. The method of claim 1, wherein: When the generation condition is met, generating a virtual lane line based on the left and right lane lines of the current lane includes: Checking whether the left and right lane lines of the current lane are identified and whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; When the left and right lane lines of the current lane are identified and the intersection function flag is valid, checking whether the preview distance of the left and right lane lines is greater than a second preset threshold; When it is determined that the left and right lane lines have not been updated in a predetermined number of consecutive image frames, generating the virtual lane lines based on historical lane lines; and Update the attributes of the virtual lane line.

6. The method of claim 1, wherein: Connecting the generated virtual lane line with the opposite lane line at the intersection includes: Checking whether the left and right lane lines of the current lane are identified and whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; When the oncoming lane line at the intersection is identified, the virtual lane line is updated using the oncoming lane line at the intersection to obtain a new virtual lane line; and When the lateral deviation between the new virtual lane line and the virtual lane line before updating is within a third preset threshold, the left and right lane lines in the new virtual lane line are made parallel to each other.

7. The method of claim 2, wherein: After the exit condition is met, exiting the virtual lane line generation method includes: Checking whether the intersection function flag is valid, wherein the intersection function flag is set to be valid when the generation condition is met; and After the intersection function flag is set to valid, when the driving distance and driving time of the vehicle exceed a fourth preset threshold, it is determined that the exit condition is met and the virtual lane line generation method is exited.

8. The method of claim 7, wherein: After the exit condition is met, exiting the virtual lane line generation method further includes: When the driving distance and driving time of the vehicle do not exceed a fourth preset threshold after the intersection function flag is set to valid, but the curb and lane lines of adjacent lanes can be stably identified, check the zebra crossing at the intersection and the lane lines of the opposite lane at the intersection; and When the zebra crossing is identified, or when the oncoming lane line at the intersection is identified and the virtual lane line is successfully updated using the oncoming lane line at the intersection, it is determined that the exit condition is met and the virtual lane line generation method is exited.

9. A computer storage medium, characterized in that The medium includes instructions that, when executed, perform the method of any one of claims 1 to 8.

10. A computer program product, comprising a computer program, characterized in that When the computer program is executed by a processor, the method according to any one of claims 1 to 8 is implemented.

11. A virtual lane line generating device, characterized in that: The device comprises: A checking device, used to check whether the current scene meets the generation conditions of the virtual lane line at the intersection; A generating device, for generating a virtual lane line based on the left and right lane lines of the current lane when the generating condition is met; and The connecting device is used to connect the generated virtual lane line with the opposite lane line of the intersection.

12. The apparatus of claim 11, further comprising: The exit device is used to close the virtual lane line generating device after the exit condition is met.

13. A camera, characterized in that: The camera comprises: A collection device for collecting data related to the current scene; and The virtual lane line generating device as described in claim 11 or 12 is used to generate a virtual lane line based on the data related to the current scene.

Images