Vehicle passing detection method and device and vehicle

Abstract

The invention relates to the field of vehicle control, in particular to a vehicle passing detection method and device and a vehicle. According to the method, the binocular vision system of the vehicle is used for carrying out feature point recognition and geometric operation on the road clearance facility, dependence on expensive sensors such as a laser radar is avoided, the hardware cost is reduced, the system carrying rate is improved, and the problems that an existing scheme is high in cost and difficult to popularize are solved; secondly, automatically calculating a limit parameter based on an adaptive size measurement strategy, and objectively comparing the limit parameter with the vehicle size, thereby replacing subjective visual judgment of a driver influenced by environment and experience, and eliminating personal errors; finally, the accurate passing information is output to assist the driver in decision making, scratching or collision accidents are effectively prevented, the problem that the judgment reliability is poor when no auxiliary system exists is solved, and the driving safety is improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle control, and more specifically to a method, device, and vehicle for detecting vehicle passage. Background Technology

[0002] With the evolution of intelligent driving technology, the requirements for the accuracy and real-time performance of vehicle environmental perception are becoming increasingly stringent. Distance measurement, as the core of perception, directly determines the safety and accuracy of decision-making. However, safe passage remains a challenge for drivers in scenarios with widespread height and width restrictions, such as ordinary roads, residential areas, and rural areas.

[0003] Existing technologies mainly rely on expensive sensors such as lidar or dedicated binocular cameras, resulting in high hardware costs, complex algorithms, and low vehicle adoption rates, making it difficult to popularize in low- and mid-range models. For vehicles without advanced driver assistance systems, drivers rely solely on subjective visual judgment, which is easily affected by lighting, angle, and experience limitations, leading to frequent accidents such as scrapes and collisions. Summary of the Invention

[0004] In view of this, embodiments of the present invention provide a method, device and vehicle for detecting vehicle passage, in order to solve the problems of high cost, low adoption rate and high implementation difficulty of current sensor ranging solutions, and poor reliability of driver's subjective judgment of height and width limits when there is no auxiliary system, which can easily lead to safety accidents.

[0005] In a first aspect, embodiments of the present invention provide a method for detecting vehicle passage, the method comprising: When there are road clearance facilities in the direction of vehicle travel, the vehicle's binocular vision system is used to identify the target feature points of the road clearance facilities. The target feature points of the road clearance facility are analyzed to obtain the clearance dimension parameters of the road clearance facility; Based on the comparison results between the clearance dimension parameters and the vehicle dimension parameters, vehicle passage information is output, wherein the vehicle passage information is used to indicate that the passage of the vehicle is not affected by the road clearance facilities.

[0006] As an optional embodiment, the analysis of target feature points of the road clearance facility to obtain the clearance dimension parameters of the road clearance facility includes: Obtain the size measurement strategy adapted to the clearance attributes of the road clearance facility; The clearance dimension parameters of the road clearance facility are obtained by calculating the target feature points according to the size measurement strategy.

[0007] As an optional embodiment, the clearance attribute includes a width limit, and the target feature points include at least two feature points of the road clearance facilities on both sides of the lane where the vehicle is located; The step of calculating the clearance dimension parameters of the road clearance facility for the target feature points according to the size measurement strategy includes: A pair of symmetrical points is determined from the at least two feature points, the pair of symmetrical points including a first feature point and a second feature point symmetrically arranged on both sides of the lane; At least one target point pair is determined from the symmetrical point pairs. Boundary analysis is performed on the first and second feature points in each target point pair to obtain the width limit sub-parameters corresponding to each target point pair. Based on the obtained width limit sub-parameters, the boundary size parameters of the road boundary facility are determined.

[0008] As an optional embodiment, the boundary analysis performed on the first and second feature points of each target point pair to obtain the width-limiting sub-parameters corresponding to each target point pair includes: Determine a first distance between the first feature point and the vehicle, and determine a second distance between the second feature point and the vehicle; Based on the first distance and the second distance, a first included angle between the first connecting line and the second connecting line is determined; the first connecting line is the line connecting the first feature point and the vehicle, and the second connecting line is the line connecting the second feature point and the vehicle. Based on the first distance, the second distance, and the first included angle, determine the width-limiting sub-parameters corresponding to the target point pair.

[0009] As an optional embodiment, the target point pair includes at least two, and the determination of the clearance dimension parameters of the road clearance facility based on the obtained width restriction sub-parameters includes: The minimum value among the width-limiting sub-parameters corresponding to each target point is determined as the clearance dimension parameter of the road clearance facility.

[0010] As an optional embodiment, the clearance attribute includes a height limit, and the target feature point includes at least one feature point of the road clearance facility on one side of the lane where the vehicle is located; The step of calculating the clearance dimension parameters of the road clearance facility for the target feature points according to the size measurement strategy includes: A third feature point characterizing the top structure of the road clearance facility is determined from the at least one feature point; The reference plane of the vehicle's driving direction is used as the projection plane, and the third feature point is projected perpendicularly onto the projection plane to obtain the projected feature point; Determine the third distance between the third feature point and the vehicle, and determine the fourth distance between the projected feature point and the vehicle; Based on the third distance and the fourth distance, the second included angle between the third line and the fourth line is determined; the third line is the line connecting the third feature point and the vehicle, and the fourth line is the line connecting the projected feature point and the vehicle. Based on the third distance, the fourth distance, and the second included angle, the height limit dimensions of the road clearance facility are determined.

[0011] As an optional embodiment, the binocular vision system includes a wide-angle camera and a monocular camera; Determining the distance between the feature point and the vehicle includes: Based on the second image captured by the monocular camera, the first image captured by the wide-angle camera is corrected to obtain a third image, wherein the coordinate system of the third image is consistent with that of the second image; Based on the pixel coordinates of the feature points in the third image and the second image respectively, the binocular three-dimensional coordinates of the feature points in the binocular coordinate system are determined, wherein the binocular coordinate system is set based on the binocular vision system; The binocular 3D coordinates are transformed to the vehicle coordinate system to obtain the vehicle 3D coordinates corresponding to the feature points. The distance between the feature point and the vehicle is calculated based on the three-dimensional coordinates of the vehicle.

[0012] As an optional embodiment, determining the stereo 3D coordinates of the feature points in a stereo coordinate system based on the pixel coordinates of the feature points in the third image and the second image respectively includes: Determine the first pixel coordinates of the feature point in the third image; and determine the second pixel coordinates of the feature point in the second image; Based on the spatial position difference, the ranging distance between the first pixel coordinates and the second pixel coordinates is calculated, wherein the spatial position difference is determined based on the position between the wide-angle camera and the monocular camera; Based on the first pixel coordinates and the ranging distance, the binocular three-dimensional coordinates of the feature point in the binocular coordinate system are determined.

[0013] As an optional embodiment, the step of outputting vehicle passage information based on the comparison result of the clearance size parameters and the vehicle size parameters includes: If the comparison result indicates that the vehicle size parameter is less than or equal to the clearance size parameter, then first traffic information is output, wherein the first traffic information indicates that the vehicle can pass through the road clearance facility; or... If the comparison result indicates that the vehicle size parameter is greater than the clearance size parameter, then it is detected whether there is a temporary passage route that bypasses the road clearance facility in the area where the vehicle is located.

[0014] As an optional embodiment, detecting whether there is a route that bypasses the road clearance facility in the area where the vehicle is located includes: Calculate the minimum turning radius required to avoid the road clearance facility based on the vehicle's position and the road clearance facility's position; The temporary communication path is generated based on the minimum turning radius, the distribution of lane lines and obstacles in the area.

[0015] As an optional embodiment, after generating the temporary communication path based on the minimum turning radius, the distribution of lane lines and obstacles in the area, the method further includes: Predict the collision risk value between the temporary passage path and the obstacle; In response to the collision risk value being lower than the collision threshold, the temporary passage path is pushed to the vehicle.

[0016] As an optional embodiment, the method further includes: If there is no temporary passage route that bypasses the road clearance facility, the vehicle's area, the clearance dimensions of the road clearance facility, and its location information are uploaded to the server so that guidance information can be sent to other vehicles in the area through the server.

[0017] Secondly, embodiments of the present invention provide a vehicle passage detection device, the device comprising: The recognition module is used to identify the target feature points of the road boundary facility when there is a road boundary facility in the direction of vehicle travel using the vehicle's binocular vision system. The analysis module is used to analyze the target feature points of the road clearance facility to obtain the clearance dimension parameters of the road clearance facility; The output module is used to output vehicle passage information based on the comparison result of the clearance size parameters and the vehicle size parameters, wherein the vehicle passage information is used to indicate that the passage of the vehicle is not affected by the road clearance facilities.

[0018] Thirdly, embodiments of the present invention provide an electronic device, including: a memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, and the processor executing the computer instructions to perform the method described in the first aspect or any corresponding embodiment thereof.

[0019] Fourthly, embodiments of the present invention provide a computer-readable storage medium storing computer instructions that cause a computer to perform the method described in the first aspect or any corresponding embodiment thereof.

[0020] Fifthly, embodiments of the present invention provide a computer program product, including computer instructions, which are used to cause a computer to perform the method described in the first aspect or any corresponding embodiment thereof.

[0021] This application utilizes the vehicle's binocular vision system to perform feature point recognition and geometric calculations on road clearance facilities, avoiding reliance on expensive sensors such as LiDAR, reducing hardware costs, and increasing system adoption rates, thus solving the problems of high cost and difficulty in widespread adoption of existing solutions. Secondly, it automatically calculates clearance parameters based on an adapted size measurement strategy and objectively compares them with vehicle dimensions, replacing the driver's subjective visual judgment influenced by the environment and experience, and eliminating human error. Finally, by outputting accurate traffic information to assist the driver's decision-making, it effectively prevents scrapes or collisions, solves the problem of poor judgment reliability without assistance systems, and improves driving safety. Attached Figure Description

[0022] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0023] Figure 1 This is a flowchart illustrating a vehicle passage detection method according to some embodiments of the present invention; Figure 2 This is a schematic diagram of a binocular vision system according to some embodiments of the present invention; Figure 3 This is a schematic diagram of a width-limited scenario according to some embodiments of the present invention; Figure 4 This is a flowchart illustrating another vehicle passage detection method according to some embodiments of the present invention; Figure 5 This is a schematic diagram of a height restriction scenario according to some embodiments of the present invention; Figure 6 This is a structural block diagram of a vehicle passage detection device according to an embodiment of the present invention; Figure 7 This is a schematic diagram of the hardware structure of a computer device according to an embodiment of the present invention. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0025] According to embodiments of the present invention, a method, apparatus, and vehicle for detecting vehicle passage are provided. It should be noted that the steps shown in the flowcharts in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowcharts, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0026] Figure 1 This is a flowchart of a vehicle passage detection method according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: Step S101: When there are road clearance facilities in the direction of vehicle travel, the vehicle's binocular vision system is used to identify the target feature points of the road clearance facilities.

[0027] In one embodiment of this application, the vehicle is equipped with a binocular vision system, which includes a wide-angle camera and a monocular camera. The wide-angle camera is positioned in front of the vehicle and has an ultra-large horizontal and vertical field of view (e.g., 170°×120°), enabling it to capture a panoramic view of the vehicle's surroundings at close range. This ensures that the vehicle can identify the boundaries of road clearance facilities on both sides of the road without blind spots at complex intersections or narrow road sections, solving the problem of narrow field of view of telephoto lenses. The monocular camera is mounted in the center of the windshield and above the wide-angle camera. It adopts a telephoto design and has a narrower field of view (e.g., 30°×18°). It has high resolution and long-distance detail capture capabilities, and can clearly identify the features of clearance facilities in front of the vehicle (e.g., at 25-35 meters).

[0028] During vehicle operation, a wide-angle camera and a monocular camera simultaneously capture video streams along the vehicle's travel direction and transmit them to the cockpit domain controller. The cockpit domain controller performs timestamp alignment and frame extraction on the video stream to reduce data transmission bandwidth, and then uploads the processed image frames to the server in real time via an encrypted channel. The server deploys a large model, which performs semantic segmentation and object detection on the processed image frames to identify whether road boundary facilities exist along the vehicle's travel direction. The identification results (including category, confidence score, and image bounding box) are then sent to the vehicle. Road boundary facilities can be width-limiting barriers or height-limiting poles.

[0029] When road clearance facilities are present in the vehicle's direction of travel, the pre-calibrated intrinsic parameter matrix and distortion coefficients of the camera are first read to perform distortion correction on the image captured by the wide-angle camera, obtaining a corrected image consistent with the coordinate system of the monocular camera, thus eliminating the barrel distortion unique to wide-angle lenses. Next, using the calibrated extrinsic parameter matrix, the coordinate systems of the two cameras are aligned to the binocular coordinate system.

[0030] Subsequently, coarse matching is performed using Transformer-based image embedding vectors to locate the approximate region of the target in the two images. Then, the Scale-Invariant Feature Transform (SIFT) algorithm is used to extract target feature points with rotation invariance and illumination invariance within the target bounding box, such as the corners of the width-limiting pier and the intersections of the height-limiting poles.

[0031] Specifically, an image feature extraction network based on the Transformer architecture is invoked to perform global feature encoding on the image frames captured by the two cameras, generating high-dimensional image embedding vectors. A matching matrix is ​​constructed by calculating the cosine similarity between the two corresponding embedding vectors, and image regions with similarity higher than a preset threshold are selected to complete the coarse matching of the target, lock the approximate location of road boundary facilities such as width-limiting piers and height-limiting poles in the two images, and output the corresponding target bounding boxes, effectively narrowing the search range for subsequent fine matching.

[0032] Based on this, the SIFT algorithm is used to extract local features from the target bounding box region obtained by coarse matching. First, a multi-scale image is generated through the difference of Gaussian pyramid to enhance the algorithm's adaptability to targets at different distances. Then, an improved corner detection threshold is used to identify corresponding target feature points for different boundary attributes such as width and height restrictions: when the boundary attribute is width restriction, at least two feature points of the road boundary facilities on both sides of the vehicle's lane are extracted, including the corners of the width restriction piers and the endpoints of the intersection lines between the top and sides; when the boundary attribute is height restriction, a third feature point representing the top structure of the road boundary facility is extracted, including the intersection of the edges of the height restriction poles and the corners of the top crossbeams, etc., which have obvious geometric features.

[0033] Step S102: Analyze the target feature points of the road clearance facility to obtain the clearance dimension parameters of the road clearance facility.

[0034] In one embodiment of this application, the target feature points of the road clearance facility are analyzed to obtain the clearance dimension parameters of the road clearance facility, including: obtaining the size measurement strategy adapted to the clearance attribute of the road clearance facility; and calculating the clearance dimension parameters of the road clearance facility for the target feature points according to the size measurement strategy.

[0035] Specifically, the clearance attributes include width limit, and the target feature points include at least two feature points of the road clearance facilities on both sides of the lane where the vehicle is located. The clearance dimensions of the road clearance facilities are calculated based on the target feature points according to the size measurement strategy, including: Step A1: Determine a pair of symmetrical points from at least two feature points. The pair of symmetrical points includes a first feature point and a second feature point symmetrically located on both sides of the lane.

[0036] Specifically, when the road clearance facility's attribute is width restriction, at least two feature points are identified and extracted from the surface of the road clearance facility (such as left and right width restriction blocks) on both sides of the vehicle's lane based on a binocular vision system. Since the width restriction facilities on the left and right sides are usually symmetrical in physical structure, a spatial symmetry pairing algorithm is used to select key feature points for subsequent width restriction sub-parameter calculations, ensuring that the selected feature points can accurately reflect the actual width restriction range of the lane.

[0037] Specifically, using the vehicle's longitudinal centerline as a reference, multiple feature points on the left side are traversed, and the horizontal distance between each feature point and the vehicle's longitudinal centerline is calculated. Then, feature points on the right side of the lane are traversed, searching for feature points whose absolute horizontal distance from a certain left-side feature point to the vehicle's longitudinal centerline is equal to or within a preset tolerance range (the preset tolerance can be set according to actual road conditions, such as ±5cm, to ensure pairing accuracy). If left and right side feature points that meet the distance conditions are found, the coordinates of these two feature points in the height and depth directions are further verified to ensure they are corresponding geometric positions on the width-limiting facility (e.g., the midpoint of the inner edge of the left pier and the midpoint of the inner edge of the right pier). If the above geometric constraint verification passes, i.e., the conditions of distance symmetry and being on the same horizontal plane are met, the feature point on the left side of the lane is marked as the first feature point, and the corresponding feature point on the right side of the lane is marked as the second feature point. Together, they constitute a symmetrical point pair for subsequent boundary analysis, providing feature point data for determining the target point pair and calculating the width-limiting sub-parameters.

[0038] Step A2: Identify at least one target point pair from the symmetrical point pairs, perform boundary analysis on the first and second feature points of each target point pair to obtain the width limit sub-parameters corresponding to each target point pair, and determine the boundary dimension parameters of the road boundary facility based on the obtained width limit sub-parameters.

[0039] Specifically, a boundary analysis is performed on the first and second feature points of each target point pair to obtain the corresponding width-limiting sub-parameters for each target point pair, including: Step A21: Determine the first distance between the first feature point and the vehicle, and determine the second distance between the second feature point and the vehicle.

[0040] It should be noted that the calculation methods for the first distance between the first feature point and the vehicle, and the second distance between the second feature point and the vehicle, are exactly the same. The specific calculation process is as follows: In this embodiment, the binocular vision system includes a wide-angle camera and a monocular camera; Determining the distance between the feature point and the vehicle includes: Step a211: Based on the second image acquired by the monocular camera, the first image acquired by the wide-angle camera is corrected to obtain a third image, wherein the coordinate system of the third image is consistent with that of the second image.

[0041] Specifically, the vehicle-mounted wide-angle camera and monocular camera simultaneously acquire a first image (acquired by the wide-angle camera) and a second image (acquired by the monocular camera) of the road environment ahead. Synchronization is achieved by aligning the timestamps to ensure the two images are acquired at the same time. Then, pre-calibrated intrinsic parameter matrix, distortion coefficients, and extrinsic parameter data (used to characterize the spatial relationship between the wide-angle and monocular cameras) of the wide-angle camera are used. The correction algorithm corresponding to Zhang's calibration method is employed to correct the distortion of the first image, eliminating barrel distortion caused by the wide-angle lens. During the correction process, perspective transformation is used to map the pixel coordinate system of the first image to an imaging plane consistent with the second image, ensuring complete unification of the focal length, optical axis direction, and pixel scale of the two images. Finally, a third image with a coordinate system matching the second image is output.

[0042] Step a212: Based on the pixel coordinates of the feature points in the third and second images respectively, determine the stereo 3D coordinates of the feature points in the stereo coordinate system, wherein the stereo coordinate system is set based on the stereo vision system.

[0043] Specifically, based on the pixel coordinates of the feature points in the third and second images respectively, the stereo 3D coordinates of the feature points in the stereo coordinate system are determined, including: Step a2141: Determine the first pixel coordinates of the feature point in the third image; and determine the second pixel coordinates of the feature point in the second image.

[0044] Specifically, based on the pixel coordinates of the feature points in the third and second images respectively, the binocular three-dimensional coordinates of the feature points in the binocular coordinate system are determined, including: identifying the feature points based on the third image and determining the first pixel coordinates of the feature points in the third image; identifying the feature points based on the second image and determining the second pixel coordinates of the feature points in the second image; calculating the ranging distance corresponding to the feature points based on the first pixel coordinates, the second pixel coordinates, and the spatial position difference, wherein the spatial position difference is the position difference between the wide-angle camera and the monocular camera; and determining the binocular three-dimensional coordinates of the feature points in the binocular coordinate system according to the first pixel coordinates and the ranging distance.

[0045] Step a2142: Calculate the distance between the coordinates of the first pixel and the coordinates of the second pixel based on the spatial position difference, wherein the spatial position difference is determined based on the position between the wide-angle camera and the monocular camera.

[0046] Specifically, in the third image, the first pixel coordinates (u1, v1) of a feature point (such as the corner of a width-limiting pier) are located; then, in the second image, corresponding points are found in the second image through feature descriptor matching, and their second pixel coordinates (u2, v2) are determined; next, the pre-calibrated spatial position difference between the wide-angle camera and the monocular camera is retrieved (refer to...). Figure 2 As shown), the horizontal parallax is calculated by combining the first pixel coordinates (u1, v2) and the second pixel coordinates (u2, v2). And the equivalent focal length, the distance between the feature point and the binocular vision system is calculated based on the principle of triangulation, as shown in the following formula: In the formula, Z is the distance measured, and B is the spatial position difference. For equivalent focal length, This refers to horizontal parallax.

[0047] Step a2143: Determine the stereo 3D coordinates of the feature point in the stereo coordinate system based on the first pixel coordinates and the distance measured.

[0048] Specifically, the coordinates of the first pixel (u1, v1) are combined with the distance Z to complete the coordinate transformation from two-dimensional image to three-dimensional space, and finally the complete three-dimensional coordinates of the feature point in the binocular coordinate system are obtained, which provides core data for the subsequent transformation of the binocular three-dimensional coordinates to the vehicle coordinate system and the calculation of the distance between the feature point and the vehicle.

[0049] Step a213: Transform the binocular 3D coordinates to the vehicle coordinate system to obtain the vehicle 3D coordinates corresponding to the feature points.

[0050] Specifically, the transformation matrix (including the rotation matrix R and translation vector T) between the binocular coordinate system and the vehicle coordinate system, obtained beforehand through vehicle calibration, is acquired. During the transformation, the binocular 3D coordinates are constructed into homogeneous coordinates, and the coordinate transformation is completed through matrix multiplication. The vehicle coordinate system has the vehicle's center of mass as its origin, the X-axis along the vehicle's forward direction, the Y-axis along the vehicle's lateral direction, and the Z-axis perpendicular to the ground and upwards. The final output is the 3D coordinates of the vehicle corresponding to the feature points.

[0051] Step a214: Calculate the distance between the feature point and the vehicle based on the three-dimensional coordinates of the whole vehicle.

[0052] Specifically, based on the obtained three-dimensional coordinates of the feature point, the X-axis represents the distance in the vehicle's forward direction, the Y-axis represents the lateral distance, and the Z-axis represents the vertical height. Since the origin of the vehicle coordinate system is the vehicle's center of mass, calculating the distance between the feature point and the vehicle is essentially calculating the Euclidean distance from that point to the origin in space. Using the Euclidean distance formula, the X, Y, and Z coordinate components of the feature point are squared respectively, the squared results are added together, and finally, the square root of the sum is taken. This calculation result is the straight-line distance between the feature point and the vehicle's center of mass.

[0053] Step A22: Based on the first distance and the second distance, determine the first included angle between the first connecting line and the second connecting line; the first connecting line is the line connecting the first feature point and the vehicle, and the second connecting line is the line connecting the second feature point and the vehicle.

[0054] Specifically, the angle between the lines connecting the two feature points and the vehicle is determined through vector geometric operations. The first line connecting the vehicle origin to the first feature point is defined as vector V1, and the second line connecting the vehicle origin to the second feature point is defined as vector V2. Then, the first angle α between these two vectors is calculated using the dot product formula. The formula is expressed as: , The numerator is the dot product of vectors V1 and V2, and the denominator is the product of the magnitudes of vectors V1 and V2. In this embodiment, the radian value of α is obtained by using the inverse cosine function and then converted into an angle value.

[0055] Step A23: Based on the first distance, the second distance, and the first included angle, determine the corresponding width-limiting sub-parameters for the target point pair.

[0056] Specifically, such as Figure 3 As shown, after obtaining the first distance x1 between the first feature point and the vehicle, the second distance x2 between the second feature point and the vehicle, and the first included angle α, the width restriction sub-parameter between the left and right clearance facilities is calculated using the triangle cosine theorem. In the triangle formed by the vehicle, the first feature point, and the second feature point, the width restriction sub-parameter is the side opposite the first included angle α in that triangle, and the calculation formula is as follows:

[0057] In the formula, W is the width-limiting sub-parameter, x1 is the first distance between the first feature point and the vehicle, x2 is the second distance between the second feature point and the vehicle, and α is the first included angle.

[0058] Specifically, the target point pair includes at least two, and the clearance dimension parameters of the road clearance facility are determined based on the obtained width restriction sub-parameters, including: determining the minimum value among the width restriction sub-parameters corresponding to each target point pair as the clearance dimension parameter of the road clearance facility.

[0059] The minimum value among the various width-limiting sub-parameters is selected as the clearance dimension parameter of the road clearance facility because the width-limiting sub-parameters are based on the actual lane width data calculated from different target point pairs. The calculation results for different target point pairs (such as symmetrical point pairs of width-limiting blocks at different heights) are different. Selecting the minimum value can ensure that there is sufficient safety margin when vehicles pass through, avoid vehicle collisions caused by calculation deviations of some target point pairs or local protrusions of the width-limiting facility, and maximize traffic safety.

[0060] Step S103: Based on the comparison results of the clearance dimension parameters and the vehicle dimension parameters, output vehicle passage information, wherein the vehicle passage information is used to indicate that the passage of the vehicle is not affected by the road clearance facilities.

[0061] In one embodiment of this application, vehicle passage information is output based on a comparison between clearance dimension parameters and vehicle dimension parameters, including the following two cases: Case 1: If the comparison result shows that the vehicle size parameter is less than or equal to the clearance size parameter, then the first passage information is output, where the first passage information indicates that the vehicle can pass through the road clearance facility.

[0062] The height restriction parameter (H) is compared with the vehicle's dimensions (such as vehicle height). If the vehicle height is less than or equal to the height restriction parameter, it is determined that the vehicle can safely pass through the road clearance facility. At this time, the cockpit domain controller generates the first passage information, displays a "Safe to pass" sign on the central control screen, or projects a virtual safe passage frame onto the road height restriction area via AR-HUD, while simultaneously announcing in voice, "Height restriction ahead X centimeters, your vehicle height X centimeters, safe to pass."

[0063] Scenario 2: If the comparison result shows that the vehicle size parameters are greater than the clearance size parameters, then check whether there is a temporary passage route that bypasses the road clearance facilities in the area where the vehicle is located.

[0064] The height restriction parameter (H) is compared with the vehicle's dimensions (such as vehicle height). If the vehicle height is greater than the height restriction parameter, the vehicle cannot safely pass through the road clearance facility. At this point, based on the vehicle's current location and a high-precision map, a search area centered on the height restriction facility is defined, and alternative routes such as side roads and auxiliary roads are extracted within this area. Inaccessible routes are eliminated by combining binocular vision recognition of on-site traffic signs (such as no U-turns and one-way streets). Simultaneously, the height restriction parameters of the alternative routes are verified, excluding routes that still contain narrow sections. If a valid temporary detour route is selected, the detour distance and time are calculated, and the route is highlighted on the navigation interface, with a pop-up prompt and voice guidance stating, "Height restriction ahead prevents passage; a detour route has been planned for you," for the driver's confirmation.

[0065] Figure 4 This is a flowchart of a vehicle passage detection method according to an embodiment of the present invention, such as... Figure 4 As shown, the process includes the following steps: Step S201: When there are road clearance facilities in the direction of vehicle travel, the vehicle's binocular vision system is used to identify the target feature points of the road clearance facilities.

[0066] In one embodiment of this application, the vehicle is equipped with a binocular vision system, which includes a wide-angle camera and a monocular camera. The wide-angle camera is positioned in front of the vehicle and has an ultra-large horizontal and vertical field of view (e.g., 170°×120°), enabling it to capture a panoramic view of the vehicle's surroundings at close range. This ensures that the vehicle can identify the boundaries of road clearance facilities on both sides of the road without blind spots at complex intersections or narrow road sections, solving the problem of narrow field of view of telephoto lenses. The monocular camera is mounted in the center of the windshield and above the wide-angle camera. It adopts a telephoto design and has a narrower field of view (e.g., 30°×18°). It has high resolution and long-distance detail capture capabilities, and can clearly identify the features of clearance facilities in front of the vehicle (e.g., at 25-35 meters).

[0067] During vehicle operation, the wide-angle camera and monocular camera simultaneously capture video streams along the vehicle's travel direction and transmit them to the cockpit domain controller. The cockpit domain controller performs timestamp alignment and frame extraction on the video stream to reduce data transmission bandwidth, and then uploads the processed image frames to the server in real time via an encrypted channel. The server deploys a large model, which performs semantic segmentation and object detection on the processed image frames to identify whether road boundary facilities exist along the vehicle's travel direction. The identification results (including category, confidence score, and image coordinate frame) are then sent to the vehicle. Road boundary facilities can be width-limiting barriers, height-limiting poles, etc.

[0068] When road clearance facilities are present in the vehicle's direction of travel, the pre-calibrated intrinsic parameter matrix and distortion coefficients of the camera are first read to perform distortion correction on the image captured by the wide-angle camera, obtaining a corrected image consistent with the coordinate system of the monocular camera, thus eliminating the barrel distortion unique to wide-angle lenses. Next, using the calibrated extrinsic parameter matrix, the coordinate systems of the two cameras are aligned to the binocular coordinate system.

[0069] Subsequently, coarse matching is performed using Transformer-based image embedding vectors to locate the approximate region of the target in the two images. Then, the Scale-Invariant Feature Transform (SIFT) algorithm is used to extract target feature points with rotation invariance and illumination invariance within the target bounding box, such as the corners of the width-limiting pier and the intersections of the height-limiting poles.

[0070] Specifically, an image feature extraction network based on the Transformer architecture is invoked to perform global feature encoding on the image frames captured by the two cameras, generating high-dimensional image embedding vectors. A matching matrix is ​​constructed by calculating the cosine similarity between the two corresponding embedding vectors, and image regions with similarity higher than a preset threshold are selected to complete the coarse matching of the target, lock the approximate location of road boundary facilities such as width-limiting piers and height-limiting poles in the two images, and output the corresponding target bounding boxes, effectively narrowing the search range for subsequent fine matching.

[0071] Based on this, the SIFT algorithm is used to extract local features from the target bounding box region obtained by coarse matching. First, a multi-scale image is generated through the difference of Gaussian pyramid to enhance the algorithm's adaptability to targets at different distances. Then, an improved corner detection threshold is used to identify corresponding target feature points for different boundary attributes such as width and height restrictions: when the boundary attribute is width restriction, at least two feature points of the road boundary facilities on both sides of the vehicle's lane are extracted, including the corners of the width restriction piers and the endpoints of the intersection lines between the top and sides; when the boundary attribute is height restriction, a third feature point representing the top structure of the road boundary facility is extracted, including the intersection of the edges of the height restriction poles and the corners of the top crossbeams, etc., which have obvious geometric features.

[0072] Step S202: Analyze the target feature points of the road clearance facility to obtain the clearance dimension parameters of the road clearance facility.

[0073] In one embodiment of this application, the target feature points of the road clearance facility are analyzed to obtain the clearance dimension parameters of the road clearance facility, including: obtaining the size measurement strategy adapted to the clearance attribute of the road clearance facility; and calculating the clearance dimension parameters of the road clearance facility for the target feature points according to the size measurement strategy.

[0074] Specifically, the clearance attributes include height restrictions, and the target feature points include at least one feature point of the road clearance facility on one side of the lane where the vehicle is located.

[0075] The clearance dimensions of the road clearance facilities are calculated based on the target feature points according to the size measurement strategy, including: Step B11: Determine a third feature point from at least one feature point that characterizes the top structure of the road clearance facility.

[0076] Specifically, when the road clearance facility's attribute is height restriction, a binocular vision system is used to identify and extract the third feature point of the top structure of the road clearance facility (such as height restriction poles or height restriction frames). To accurately obtain the height restriction dimensions, key points representing the top boundary of the facility need to be selected. Specifically, the Z-axis coordinates (height values) of all feature points in the vehicle coordinate system are traversed, their statistical distribution is calculated, and low elevation points caused by ground noise or texture at the bottom of the pole are eliminated. Subsequently, feature points with the largest Z-axis coordinate values ​​or those within a preset height range are selected. These feature points typically correspond to the lower edge of the crossbeam at the top of the height restriction pole, the top of the sign bracket, or the apex of the pole. The feature points used to characterize the top structure of the height restriction facility are marked as third feature points, and their spatial position directly determines the vertical height at which vehicles can pass.

[0077] Step B12: Use the reference plane of the vehicle's driving direction as the projection plane, and project the third feature point perpendicularly onto the projection plane to obtain the projected feature point.

[0078] Specifically, to construct the right-angled triangle geometric model for calculating height, the vertical plane containing the vehicle's direction of travel (i.e., the XZ plane of the vehicle coordinate system, Y=0) is defined as the projection plane. This plane represents the theoretical centerline position of the vehicle when passing through the height restriction facility. Next, a spatial projection transformation operation is performed on the third feature point. That is, the coordinate values ​​of the third feature point on the X-axis (longitudinal distance) and Z-axis (height) remain unchanged, while its coordinate value on the Y-axis (lateral offset) is forcibly set to zero. This is equivalent to drawing a perpendicular line from the third feature point to the reference plane in the vehicle's direction of travel; the intersection of this perpendicular line and the reference plane is the projected feature point.

[0079] Step B13: Determine the third distance between the third feature point and the vehicle, and determine the fourth distance between the projected feature point and the vehicle.

[0080] It should be noted that the calculation methods for the third distance between the third feature point and the vehicle, and the fourth distance between the projected feature point and the vehicle, are exactly the same. The specific calculation process is as follows: In this embodiment, the binocular vision system includes a wide-angle camera and a monocular camera; Determining the distance between the feature point and the vehicle includes: Step b211: Based on the second image captured by the monocular camera, the first image captured by the wide-angle camera is corrected to obtain a third image, wherein the coordinate system of the third image is consistent with that of the second image.

[0081] Specifically, the vehicle-mounted wide-angle camera and monocular camera simultaneously acquire a first image (acquired by the wide-angle camera) and a second image (acquired by the monocular camera) of the road environment ahead. Synchronization is achieved by aligning the timestamps to ensure the two images are acquired at the same time. Then, pre-calibrated intrinsic parameter matrix, distortion coefficients, and extrinsic parameter data (used to characterize the spatial relationship between the wide-angle and monocular cameras) of the wide-angle camera are used. The correction algorithm corresponding to Zhang's calibration method is employed to correct the distortion of the first image, eliminating barrel distortion caused by the wide-angle lens. During the correction process, perspective transformation is used to map the pixel coordinate system of the first image to an imaging plane consistent with the second image, ensuring complete unification of the focal length, optical axis direction, and pixel scale of the two images. Finally, a third image with a coordinate system matching the second image is output.

[0082] Step b212: Based on the pixel coordinates of the feature points in the third and second images respectively, determine the stereo 3D coordinates of the feature points in the stereo coordinate system, wherein the stereo coordinate system is set based on the stereo vision system.

[0083] Specifically, based on the pixel coordinates of the feature points in the third and second images respectively, the stereo 3D coordinates of the feature points in the stereo coordinate system are determined, including: Step b2141: Determine the first pixel coordinates of the feature point in the third image; and determine the second pixel coordinates of the feature point in the second image.

[0084] Specifically, based on the pixel coordinates of the feature points in the third and second images respectively, the binocular three-dimensional coordinates of the feature points in the binocular coordinate system are determined, including: identifying the feature points based on the third image and determining the first pixel coordinates of the feature points in the third image; identifying the feature points based on the second image and determining the second pixel coordinates of the feature points in the second image; calculating the ranging distance corresponding to the feature points based on the first pixel coordinates, the second pixel coordinates, and the spatial position difference, wherein the spatial position difference is the position difference between the wide-angle camera and the monocular camera; and determining the binocular three-dimensional coordinates of the feature points in the binocular coordinate system according to the first pixel coordinates and the ranging distance.

[0085] Step b2142: Calculate the distance between the coordinates of the first pixel and the coordinates of the second pixel based on the spatial position difference, wherein the spatial position difference is determined based on the position between the wide-angle camera and the monocular camera.

[0086] Specifically, in the third image, the first pixel coordinates (u1, v1) of a feature point (such as the corner of a width-limiting pier) are located; then, in the second image, corresponding points are found in the second image through feature descriptor matching, and their second pixel coordinates (u2, v2) are determined; next, the pre-calibrated spatial position difference between the wide-angle camera and the monocular camera is retrieved (refer to...). Figure 2 As shown), the horizontal parallax is calculated by combining the first pixel coordinates (u1, v2) and the second pixel coordinates (u2, v2). And the equivalent focal length, the distance between the feature point and the binocular vision system is calculated based on the principle of triangulation, as shown in the following formula: In the formula, Z is the distance measured, and B is the spatial position difference. For equivalent focal length, This refers to horizontal parallax.

[0087] Step b2143: Determine the stereo 3D coordinates of the feature point in the stereo coordinate system based on the first pixel coordinates and the distance measured.

[0088] Specifically, the coordinates of the first pixel (u1, v1) are combined with the distance Z to complete the coordinate transformation from two-dimensional image to three-dimensional space, and finally the complete three-dimensional coordinates of the feature point in the binocular coordinate system are obtained, which provides core data for the subsequent transformation of the binocular three-dimensional coordinates to the vehicle coordinate system and the calculation of the distance between the feature point and the vehicle.

[0089] Step b213: Transform the binocular 3D coordinates to the vehicle coordinate system to obtain the vehicle 3D coordinates corresponding to the feature points.

[0090] Specifically, the transformation matrix (including the rotation matrix R and translation vector T) between the binocular coordinate system and the vehicle coordinate system, obtained beforehand through vehicle calibration, is acquired. During the transformation, the binocular 3D coordinates are constructed into homogeneous coordinates, and the coordinate transformation is completed through matrix multiplication. The vehicle coordinate system has the vehicle's center of mass as its origin, the X-axis along the vehicle's forward direction, the Y-axis along the vehicle's lateral direction, and the Z-axis perpendicular to the ground and upwards. The final output is the 3D coordinates of the vehicle corresponding to the feature points.

[0091] Step b214: Calculate the distance between the feature point and the vehicle based on the three-dimensional coordinates of the whole vehicle.

[0092] Specifically, based on the obtained three-dimensional coordinates of the feature point, the X-axis represents the distance in the vehicle's forward direction, the Y-axis represents the lateral distance, and the Z-axis represents the vertical height. Since the origin of the vehicle coordinate system is the vehicle's center of mass, calculating the distance between the feature point and the vehicle is essentially calculating the Euclidean distance from that point to the origin in space. Using the Euclidean distance formula, the X, Y, and Z coordinate components of the feature point are squared respectively, the squared results are added together, and finally, the square root of the sum is taken. This calculation result is the straight-line distance between the feature point and the vehicle's center of mass.

[0093] Step B14: Based on the third distance and the fourth distance, determine the second included angle between the third line and the fourth line; the third line is the line connecting the third feature point and the vehicle, and the fourth line is the line connecting the projected feature point and the vehicle.

[0094] Specifically, the angle between the lines connecting these two feature points and the vehicle is determined through vector geometric operations. The third line connecting the vehicle origin to the third feature point is defined as vector V3, and the fourth line connecting the vehicle origin to the projected feature point is defined as vector V4. Then, the second angle α between these two vectors is calculated using the dot product formula. The formula is expressed as: , The numerator is the dot product of vectors V3 and V4, and the denominator is the product of the magnitudes of vectors V3 and V4. In this embodiment, the radian value of α is calculated using the inverse cosine function and then converted into an angle value.

[0095] Step B15: Determine the height limit parameters of the road clearance facility based on the third distance, the fourth distance, and the second included angle.

[0096] Specifically, such as Figure 5 As shown, after obtaining the third distance x3 between the third feature point and the vehicle, the fourth distance x4 between the projected feature point and the vehicle, and the second included angle α, the height restriction dimension parameters are calculated using the triangle cosine theorem. In the triangle formed by the vehicle, the third feature point, and the projected feature point, the width restriction dimension parameter is the opposite side of the second included angle α in that triangle, and the calculation formula is as follows:

[0097] In the formula, H is the height limit dimension parameter, x3 is the third distance between the third feature point and the vehicle, and α is the second included angle.

[0098] Step S203: Based on the comparison results of the clearance size parameters and the vehicle size parameters, output vehicle passage information, wherein the vehicle passage information is used to indicate that the passage of vehicles is not affected by road clearance facilities.

[0099] In one embodiment of this application, vehicle passage information is output based on a comparison between clearance dimension parameters and vehicle dimension parameters, including the following two cases: Case 1: If the comparison result shows that the vehicle size parameter is less than or equal to the clearance size parameter, then the first passage information is output, where the first passage information indicates that the vehicle can pass through the road clearance facility.

[0100] The height restriction parameter (H) is compared with the vehicle size parameter (such as the vehicle height parameter). If the vehicle height parameter is smaller than the height restriction parameter, the difference between the height restriction parameter and the vehicle height parameter is calculated. If this difference is greater than or equal to the safe height threshold, it is determined that the vehicle can safely pass through the road clearance facility. At this time, the cockpit domain controller generates the first passage information, which can be displayed as a "Safe to pass" sign on the central control screen, or a virtual safe passage frame can be projected onto the road height restriction area via AR-HUD, while simultaneously announcing in voice, "Height restriction ahead X centimeters, your vehicle height X centimeters, safe to pass."

[0101] Scenario 2: If the comparison result shows that the vehicle size parameters are greater than the clearance size parameters, then check whether there is a temporary passage route that bypasses the road clearance facilities in the area where the vehicle is located.

[0102] The height restriction parameter (H) is compared with the vehicle size parameter (such as the vehicle height parameter). If the vehicle height parameter is greater than or equal to the height restriction parameter, or if the vehicle height parameter is less than the height restriction parameter and the difference between the height restriction parameter and the vehicle height parameter is less than the safe height threshold, then it is determined that the vehicle cannot safely pass through the road clearance facility. At this time, based on the vehicle's current area and a high-precision map, a search range centered on the height restriction facility is defined, and alternative routes such as side roads and auxiliary roads within the range are extracted. Inaccessible routes are eliminated by combining binocular visual recognition of on-site traffic signs (such as no U-turns and one-way streets), while simultaneously verifying the height restriction parameters of alternative routes to exclude routes that still have narrow sections. If a valid temporary detour route is selected, the detour distance and time are calculated, and the route is presented as a highlighted route on the navigation interface, with a pop-up prompt and voice guidance stating "Height restriction ahead prevents passage; a detour route has been planned for you," for the driver's confirmation.

[0103] In one embodiment of this application, detecting whether there is a route that bypasses road clearance facilities in the area where the vehicle is located includes: Step C1: Calculate the minimum turning radius required to avoid the road clearance facility based on the vehicle's position and the location of the road clearance facility.

[0104] Specifically, firstly, a high-precision map is retrieved, and combined with the vehicle's current positioning information, the vehicle's real-time position coordinates are determined. Simultaneously, the coordinates of feature points of road clearance facilities calculated by a binocular vision system are used to determine the location of obstacles. The lateral distance (i.e., deviation distance) and longitudinal distance between the vehicle's center of gravity and the center point of the clearance facility are calculated. Based on the Ackerman steering geometry model, the vehicle's wheelbase parameters, and the current maximum steering angle limit, the minimum turning radius required for the vehicle to complete the turn is calculated.

[0105] Then, based on the relative position of the vehicle and the obstacle, it is determined whether the vehicle needs to detour to the left or right.

[0106] Step C2: Based on the minimum turning radius, the distribution of lane lines and obstacles in the area, a temporary communication path is generated.

[0107] Specifically, based on the minimum turning radius, the distribution of lane lines and obstacles in the area, a temporary communication path is generated, including: identifying the distribution of lane lines and obstacles in the area, determining whether the vehicle has the physical space to change lanes in the direction of the minimum turning radius; obtaining the connecting roads in the corresponding direction of the physical space, and generating a temporary passage path.

[0108] Specifically, the process involves identifying lane markings in the current lane and adjacent lanes to determine the physical boundaries of the road (such as shoulders and guardrails). Next, it involves constructing the envelope of the vehicle's minimum turning radius trajectory, which is the maximum contour range traversed by the vehicle's four corner points during the turning process.

[0109] Simultaneously, the system detects whether there are dynamic obstacles (such as moving vehicles or pedestrians) or static obstacles (such as trees or utility poles) within the vehicle's steering envelope. By comparing the spatial coordinates of the vehicle's steering envelope with those of the environmental obstacles, it determines whether there is sufficient clearance in the target direction for the vehicle to complete the steering maneuver. If there are no obstacles within the envelope and the vehicle does not exceed the physical boundaries of the road, it is determined that the vehicle has the physical space to change lanes in the direction of the minimum turning radius.

[0110] If the physical space is confirmed to exist, the topology of the high-precision map is queried to find roads that connect the vehicle's current location in the turning direction, such as side roads, auxiliary roads, or U-turns. A smooth temporary route is generated on these connecting roads. This route consists of a series of discrete waypoints. The route data is converted from the map coordinate system to the vehicle's local coordinate system and encapsulated into navigation command data. This data includes the curvature information of the route, suggested speed, and steering instructions. The temporary route is then displayed to the driver via the in-vehicle infotainment system as a highlighted line on the central control screen or AR-HUD, accompanied by a voice prompt: "The road ahead is impassable. A temporary detour has been planned for you. Please follow the navigation."

[0111] Specifically, after generating a temporary communication path based on the minimum turning radius, the distribution of lane lines and obstacles in the area, the process also includes: predicting the collision risk value between the temporary travel path and obstacles; and pushing the temporary travel path to the vehicle in response to the collision risk value being lower than the collision threshold.

[0112] The temporary passage path is projected onto a real-time obstacle detection map, and the distance between each sampling point on the path and surrounding obstacles is calculated. Simultaneously, combined with the obstacle's motion vectors (velocity and direction), the position of obstacles in the next few seconds is predicted, and it is determined whether the vehicle will spatiotemporally overlap with obstacles while traveling along the path. If all points on the path remain within a safe distance threshold, a collision risk is predicted. In response to a collision risk value lower than the collision threshold, a temporary passage path is pushed to the vehicle. The collision threshold is a pre-calibrated safety threshold (dynamically adjusted based on vehicle size and speed). When the collision risk is lower than this threshold, it indicates that the temporary passage path's safety meets the vehicle's detour requirements, and the path is pushed to the vehicle, including path trajectory, turning prompts, and driving speed suggestions, ensuring that the driver or the vehicle's autonomous driving system can clearly obtain detour guidance and successfully complete the detour. If the collision risk is higher than or equal to the collision threshold, the path is not pushed, and a new temporary passage path must be generated and collision risk detection performed again.

[0113] In one embodiment of this application, the method further includes: if there is no temporary passage route around the road clearance facility, uploading the vehicle's area, the clearance size data of the road clearance facility, and the location information to the server so as to send guidance information to other vehicles in the area through the server.

[0114] Specifically, the vehicle-mounted communication unit standardizes and encapsulates the collected key information to generate a "road closure event" message. This message contains three core data fields: first, the vehicle's region, namely the vehicle's current high-precision latitude and longitude coordinates and the geofence data of a preset surrounding range (e.g., a radius of 500 meters), used to define the affected traffic area; second, the clearance dimension data, namely the width (W) or height (H) limit of the road clearance facility, and the geometric shape parameters of the road clearance facility; and third, the location information, namely the location coordinates of the road clearance facility in an absolute geographic coordinate system. Subsequently, the vehicle-mounted communication unit uploads this message to the server-side traffic management platform in real time via the mobile network. After receiving this data, the server integrates it with a high-precision digital map, updates the traffic attribute status of the road segment, and calculates other vehicles affected.

[0115] Finally, the server sends guidance information to other vehicles located in or about to enter the area via a broadcast channel. This information includes the specific location and size restrictions of the clearance facility, as well as suggested detour routes, thereby dynamically guiding traffic flow within the area and preventing other vehicles (especially large vehicles) from mistakenly entering the impassable section.

[0116] This embodiment also provides a vehicle passage detection device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0117] This embodiment provides a vehicle passage detection device, such as... Figure 6 As shown, it includes: The recognition module 601 is used to identify the target feature points of the road clearance facility when there is a road clearance facility in the direction of vehicle travel using the vehicle's binocular vision system. Analysis module 602 is used to analyze the target feature points of road clearance facilities to obtain the clearance dimension parameters of the road clearance facilities; The output module 603 is used to output vehicle passage information based on the comparison results of the clearance size parameters and the vehicle size parameters. The vehicle passage information is used to indicate that the passage of vehicles is not affected by the road clearance facilities.

[0118] In one embodiment of this application, the analysis module 602 is specifically used to obtain the size measurement strategy for the clearance attribute adaptation of the road clearance facility; and to calculate the clearance size parameters of the road clearance facility according to the size measurement strategy for the target feature points.

[0119] In one embodiment of this application, the analysis module 602 is specifically used to determine a pair of symmetrical points from at least two feature points when the clearance attribute includes width limit and the target feature points include at least two feature points of the road clearance facilities on both sides of the lane where the vehicle is located; the pair of symmetrical points includes a first feature point and a second feature point symmetrically arranged on both sides of the lane; determine at least one pair of target points from the pair of symmetrical points; perform clearance analysis on the first feature point and the second feature point in each pair of target points to obtain the width limit sub-parameters corresponding to each pair of target points; and determine the clearance dimension parameters of the road clearance facilities based on the obtained width limit sub-parameters.

[0120] In one embodiment of this application, the analysis module 602 is specifically used to determine a first distance between a first feature point and a vehicle, and to determine a second distance between a second feature point and a vehicle; based on the first distance and the second distance, to determine a first included angle between a first connecting line and a second connecting line; the first connecting line is the line connecting the first feature point and the vehicle, and the second connecting line is the line connecting the second feature point and the vehicle; based on the first distance, the second distance, and the first included angle, to determine the width limiting sub-parameters corresponding to the target point pair.

[0121] In one embodiment of this application, the analysis module 602 is specifically used to determine the minimum value among the width restriction sub-parameters corresponding to each target point pair as the clearance dimension parameter of the road clearance facility when the target point pair includes at least two.

[0122] In one embodiment of this application, the analysis module 602 is specifically used to: determine a third feature point characterizing the top structure of the road clearance facility from at least one feature point when the clearance attribute includes a height limit and the target feature point includes at least one feature point of the road clearance facility on one side of the lane where the vehicle is located; use a reference plane of the vehicle's driving direction as a projection plane and project the third feature point perpendicularly onto the projection plane to obtain a projected feature point; determine a third distance between the third feature point and the vehicle, and determine a fourth distance between the projected feature point and the vehicle; determine a second included angle between the third and fourth lines based on the third and fourth distances; the third line is the line connecting the third feature point and the vehicle, and the fourth line is the line connecting the projected feature point and the vehicle; and determine the height limit dimension parameters of the road clearance facility based on the third distance, the fourth distance, and the second included angle.

[0123] In one embodiment of this application, the binocular vision system includes a wide-angle camera and a monocular camera; The analysis module 602 is specifically used to correct the first image captured by the wide-angle camera based on the second image captured by the monocular camera to obtain a third image, wherein the coordinate system of the third image is consistent with that of the second image; based on the pixel coordinates of the feature points in the third image and the second image respectively, the binocular three-dimensional coordinates of the feature points in the binocular coordinate system are determined, wherein the binocular coordinate system is set based on the binocular vision system; the binocular three-dimensional coordinates are transformed to the vehicle coordinate system in which the vehicle is located to obtain the vehicle three-dimensional coordinates corresponding to the feature points; and the distance between the feature points and the vehicle is calculated based on the vehicle three-dimensional coordinates.

[0124] The analysis module 602 is specifically used to determine the first pixel coordinates of the feature point in the third image; and to determine the second pixel coordinates of the feature point in the second image; to calculate the distance between the first pixel coordinates and the second pixel coordinates based on the spatial position difference, wherein the spatial position difference is determined based on the position between the wide-angle camera and the monocular camera; and to determine the binocular three-dimensional coordinates of the feature point in the binocular coordinate system according to the first pixel coordinates and the distance.

[0125] In one embodiment of this application, the output module 603 is specifically used to output first passage information if the comparison result is that the vehicle size parameter is less than or equal to the clearance size parameter, wherein the first passage information indicates that the vehicle can pass through the road clearance facility; or, if the comparison result is that the vehicle size parameter is greater than the clearance size parameter, to detect whether there is a temporary passage path around the road clearance facility in the area where the vehicle is located.

[0126] In one embodiment of this application, the output module 603 is specifically used to calculate the minimum turning radius required to avoid the road clearance facility based on the vehicle's position and the location of the road clearance facility; and to generate a temporary communication path based on the minimum turning radius, the distribution of lane lines and obstacles in the area.

[0127] In one embodiment of this application, the device further includes: a push module, configured to predict a collision risk value between a temporary passage path and an obstacle; and to push a temporary passage path to the vehicle in response to a collision risk value being lower than a collision threshold.

[0128] Please see Figure 7 , Figure 7 This is a schematic diagram of the structure of a computer device provided in an optional embodiment of the present invention, such as... Figure 7 As shown, the computer device includes one or more processors 10, memory 20, and interfaces for connecting the components, including high-speed interfaces and low-speed interfaces. The components communicate with each other via different buses and can be mounted on a common motherboard or otherwise installed as needed. The processors can process instructions executed within the computer device, including instructions stored in or on memory to display graphical information of a GUI on external input / output devices (such as display devices coupled to the interfaces). In some alternative implementations, multiple processors and / or multiple buses can be used with multiple memories and multiple memory modules, if desired. Similarly, multiple computer devices can be connected, each providing some of the necessary operations (e.g., as a server array, a group of blade servers, or a multiprocessor system).

[0129] Processor 10 may be a central processing unit, a network processor, or a combination thereof. Processor 10 may further include a hardware chip. The hardware chip may be an application-specific integrated circuit (ASIC), a programmable logic device (PLD), or a combination thereof. The programmable logic device may be a complex programmable logic device (CAMP), a field-programmable gate array (FPGA), a general-purpose array logic (GDA), or any combination thereof.

[0130] The memory 20 stores instructions executable by at least one processor 10 to cause at least one processor 10 to perform the method shown in the above embodiments.

[0131] The memory 20 may include a program storage area and a data storage area. The program storage area may store the operating system and applications required for at least one function; the data storage area may store data created based on the use of the computer device as shown by a landing page for an app. Furthermore, the memory 20 may include high-speed random access memory and may also include non-transitory memory, such as at least one disk storage device, flash memory device, or other non-transitory solid-state storage device. In some alternative embodiments, the memory 20 may optionally include memory remotely located relative to the processor 10, which can be connected to the computer device via a network. Examples of such networks include, but are not limited to, the Internet, intranets, local area networks, mobile communication networks, and combinations thereof.

[0132] The memory 20 may include volatile memory, such as random access memory; the memory may also include non-volatile memory, such as flash memory, hard disk or solid-state drive; the memory 20 may also include a combination of the above types of memory.

[0133] The computer device also includes a communication interface 30 for communicating with other devices or communication networks.

[0134] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that computers, processors, microprocessor controllers, or programmable hardware include storage components capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods shown in the above embodiments.

[0135] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and all such modifications and variations fall within the scope defined by the appended claims.

Claims

1. A method for detecting vehicle passage, characterized in that, The method includes: When there are road clearance facilities in the direction of vehicle travel, the vehicle's binocular vision system is used to identify the target feature points of the road clearance facilities. The target feature points of the road clearance facility are analyzed to obtain the clearance dimension parameters of the road clearance facility; Based on the comparison results between the clearance dimension parameters and the vehicle dimension parameters, vehicle passage information is output, wherein the vehicle passage information is used to indicate that the passage of the vehicle is not affected by the road clearance facilities.

2. The method according to claim 1, characterized in that, The analysis of target feature points of the road clearance facility to obtain the clearance dimension parameters of the road clearance facility includes: Obtain the size measurement strategy adapted to the clearance attributes of the road clearance facility; The clearance dimension parameters of the road clearance facility are obtained by calculating the target feature points according to the size measurement strategy.

3. The method according to claim 2, characterized in that, The clearance attribute includes width limit, and the target feature points include at least two feature points of the road clearance facilities on both sides of the lane where the vehicle is located. The step of calculating the clearance dimension parameters of the road clearance facility for the target feature points according to the size measurement strategy includes: A pair of symmetrical points is determined from the at least two feature points, the pair of symmetrical points including a first feature point and a second feature point symmetrically arranged on both sides of the lane; At least one target point pair is determined from the symmetrical point pairs. Boundary analysis is performed on the first and second feature points in each target point pair to obtain the width limit sub-parameters corresponding to each target point pair. Based on the obtained width limit sub-parameters, the boundary size parameters of the road boundary facility are determined.

4. The method according to claim 3, characterized in that, The boundary analysis is performed on the first and second feature points of each target point pair to obtain the width-limiting sub-parameters corresponding to each target point pair, including: Determine a first distance between the first feature point and the vehicle, and determine a second distance between the second feature point and the vehicle; Based on the first distance and the second distance, a first included angle between the first connecting line and the second connecting line is determined; the first connecting line is the line connecting the first feature point and the vehicle, and the second connecting line is the line connecting the second feature point and the vehicle. Based on the first distance, the second distance, and the first included angle, determine the width-limiting sub-parameters corresponding to the target point pair.

5. The method according to claim 3, characterized in that, The target point pair includes at least two, and the determination of the clearance dimension parameters of the road clearance facility based on the obtained width restriction sub-parameters includes: The minimum value among the width-limiting sub-parameters corresponding to each target point is determined as the clearance dimension parameter of the road clearance facility.

6. The method according to claim 2, characterized in that, The clearance attribute includes a height limit, and the target feature point includes at least one feature point of the road clearance facility on one side of the lane where the vehicle is located. The step of calculating the clearance dimension parameters of the road clearance facility for the target feature points according to the size measurement strategy includes: A third feature point characterizing the top structure of the road clearance facility is determined from the at least one feature point; The reference plane of the vehicle's driving direction is used as the projection plane, and the third feature point is projected perpendicularly onto the projection plane to obtain the projected feature point; Determine the third distance between the third feature point and the vehicle, and determine the fourth distance between the projected feature point and the vehicle; Based on the third distance and the fourth distance, the second included angle between the third line and the fourth line is determined; the third line is the line connecting the third feature point and the vehicle, and the fourth line is the line connecting the projected feature point and the vehicle. Based on the third distance, the fourth distance, and the second included angle, the height limit dimensions of the road clearance facility are determined.

7. The method according to claim 3 or 6, characterized in that, The binocular vision system includes a wide-angle camera and a monocular camera; Determining the distance between the feature point and the vehicle includes: Based on the second image captured by the monocular camera, the first image captured by the wide-angle camera is corrected to obtain a third image, wherein the coordinate system of the third image is consistent with that of the second image; Based on the pixel coordinates of the feature points in the third image and the second image respectively, the binocular three-dimensional coordinates of the feature points in the binocular coordinate system are determined, wherein the binocular coordinate system is set based on the binocular vision system; The binocular 3D coordinates are transformed to the vehicle coordinate system to obtain the vehicle 3D coordinates corresponding to the feature points. The distance between the feature point and the vehicle is calculated based on the three-dimensional coordinates of the vehicle.

8. The method according to claim 7, characterized in that, Determining the stereo 3D coordinates of the feature points in the stereo coordinate system based on the pixel coordinates of the feature points in the third image and the second image respectively includes: Determine the first pixel coordinates of the feature point in the third image; and determine the second pixel coordinates of the feature point in the second image; Based on the spatial position difference, the ranging distance between the first pixel coordinates and the second pixel coordinates is calculated, wherein the spatial position difference is determined based on the position between the wide-angle camera and the monocular camera; Based on the first pixel coordinates and the ranging distance, the binocular three-dimensional coordinates of the feature point in the binocular coordinate system are determined.

9. The method according to claim 1, characterized in that, Based on the comparison result between the clearance dimension parameters and the vehicle dimension parameters, the vehicle passage information is output, including: If the comparison result indicates that the vehicle size parameter is less than or equal to the clearance size parameter, then first traffic information is output, wherein the first traffic information indicates that the vehicle can pass through the road clearance facility; or... If the comparison result indicates that the vehicle size parameter is greater than the clearance size parameter, then it is detected whether there is a temporary passage route that bypasses the road clearance facility in the area where the vehicle is located.

10. The method according to claim 9, characterized in that, Whether the area where the detection vehicle is located has a route that bypasses the road clearance facility includes: Calculate the minimum turning radius required to avoid the road clearance facility based on the vehicle's position and the road clearance facility's position; The temporary communication path is generated based on the minimum turning radius, the distribution of lane lines and obstacles in the area.

11. The method according to claim 10, characterized in that, After generating the temporary communication path based on the minimum turning radius and the distribution of lane lines and obstacles in the area, the method further includes: Predict the collision risk value between the temporary passage path and the obstacle; In response to the collision risk value being lower than the collision threshold, the temporary passage path is pushed to the vehicle.

12. The method according to claim 9, characterized in that, The method further includes: If there is no temporary passage route that bypasses the road clearance facility, the vehicle's area, the clearance dimensions of the road clearance facility, and its location information are uploaded to the server so that guidance information can be sent to other vehicles in the area through the server.

13. An electronic device, characterized in that, include: A memory and a processor, the memory and the processor being communicatively connected to each other, the memory storing computer instructions, the processor executing the computer instructions to perform the method of any one of claims 1 to 12.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions for causing a computer to perform the method of any one of claims 1 to 12.

Images