A road marking drawing method and system based on image recognition

Abstract

The application relates to the technical field of road marking, in particular to a road marking drawing method and system based on image recognition. The method comprises the following steps: acquiring a historical trajectory data set, a road design parameter set and an environment image set of a target road; analyzing the historical trajectory data set to determine a trajectory repetition degree; determining a geometric constraint condition according to the road design parameter set; determining a dangerous source distribution feature according to the environment image set; and determining a marking scheme according to the trajectory repetition degree, the geometric constraint condition and the dangerous source distribution feature. The method is helpful for improving the marking visible distance under extreme conditions, reducing out-of-control accidents caused by overspeed, reducing the contradiction between the danger source avoidance space and the road side, generating the marking scheme, avoiding that the compensation strength exceeds the physical limit of the lane, and reducing the decision error rate at the marking mutation.

Description

Technical Field

[0001] This application relates to the field of road marking technology, and in particular to a road marking drawing method and system based on image recognition. Background Technology

[0002] As a core element of traffic guidance systems, road markings directly impact driving safety and traffic efficiency. Traditional methods are used to draw markings on mountainous roads that include continuous curves, steep slopes, and high-risk roadside environments.

[0003] However, the accident-inducing rate of existing road marking schemes is higher than that of ordinary road sections. Therefore, how to solve the significant technical defects of the current road marking drawing method in the application of complex terrain road sections has become a major core issue. Summary of the Invention

[0004] This application provides a road marking drawing method and system based on image recognition to solve the above problems.

[0005] In a first aspect, this application provides a road marking drawing method based on image recognition. The method includes: acquiring a historical trajectory dataset, a road design parameter set, and an environmental image set of a target road; analyzing the historical trajectory dataset to determine the trajectory repeatability; determining geometric constraints based on the road design parameter set; determining hazard source distribution characteristics based on the environmental image set; and determining a marking scheme based on the trajectory repeatability, the geometric constraints, and the hazard source distribution characteristics.

[0006] This solution improves trajectory classification accuracy by extracting historical vehicle trajectory datasets. Extracting road design parameter sets enhances the separation of vehicle type-related trajectory features. Extracting environmental image sets provides data support for hazard distribution characteristics, compensating for the shortcomings of traditional static identification. Analyzing historical trajectory datasets helps eliminate GPS drift and abnormal driving behavior data, effectively reducing trajectory range extraction errors. Calculating trajectory repeatability improves the warning effect of lane markings and reduces critical danger states caused by insufficient safety redundancy due to trajectory competition between small vehicles and articulated vehicles. Generating geometric constraints helps reduce the rate of skid accidents and judgment errors. Determining hazard distribution characteristics improves the visibility distance of lane markings under extreme conditions, reduces runaway accidents caused by speeding, and reduces conflicts in roadside hazard avoidance space. Generating lane marking schemes helps avoid compensation intensity exceeding lane physical limitations and reduces decision-making error rates at abrupt changes in lane markings.

[0007] Optionally, the step of analyzing the historical trajectory dataset to determine the trajectory repetition includes: analyzing the historical trajectory dataset to determine the types of vehicles and the corresponding sets of driving trajectories; for each type of vehicle, analyzing the sets of driving trajectories to determine the spatial deviation distribution of the corresponding vehicle type; establishing a spatiotemporal four-dimensional filtering model; based on the spatiotemporal four-dimensional filtering model, removing abnormal data according to the spatial deviation distribution, and determining the trajectory range of the removed driving trajectory sets according to the spatial deviation distribution; and determining the trajectory repetition based on the trajectory range.

[0008] Optionally, the vehicle types include articulated vehicles and small vehicles. Determining the trajectory range of the eliminated trajectory set based on the spatial deviation distribution includes: for the articulated vehicles, analyzing the trajectory sets of the articulated vehicles and the trajectory sets of the small vehicles to obtain a first vehicle analysis result for the articulated vehicles and a second vehicle analysis result for the small vehicles; determining the turning envelope of each articulated vehicle based on the first vehicle analysis result; determining the turning boundary of the articulated vehicles based on the turning envelope; and determining the trajectory range of the eliminated trajectory set based on the turning boundary, the second vehicle analysis result, and the spatial deviation distribution.

[0009] Optionally, determining the geometric constraints based on the road design parameter set includes: parsing the road design parameter set to obtain the curve radius, longitudinal slope, and superelevation; analyzing the spatial deviation distribution of the turning envelope and small vehicles to determine the dynamic width requirement; correcting the dynamic width requirement based on the coupling relationship between the longitudinal slope and the superelevation to obtain the actual dynamic width requirement; and determining the actual dynamic width requirement as the geometric constraint.

[0010] Optionally, determining the hazard distribution characteristics based on the environmental image set includes: analyzing the environmental image set to determine road information and meteorological information; determining the distance and height difference between roadside cliffs based on the road information; determining the annual average frequency of foggy days and the duration of a single fog event based on the meteorological information; determining the fog type based on the meteorological information; determining the driving impact of each type of fog based on the fog type and the meteorological information; determining high-risk area characteristics based on the distance and height difference between roadside cliffs; determining visibility hazard characteristics based on the annual average frequency of foggy days, the driving impact, and the duration of a single fog event; and determining the high-risk area characteristics and the visibility hazard characteristics as hazard distribution characteristics.

[0011] Optionally, determining the turning envelope of each articulated vehicle based on the first vehicle analysis results includes: determining the rear axle offset data of the articulated vehicle based on the first vehicle analysis results; predicting the vehicle load status based on the rear axle offset data; determining the dynamic envelope range based on the vehicle load status and the rear axle offset data; and correcting the dynamic envelope range based on the coupling relationship between the rear axle offset data and the superelevation cross slope to obtain the turning envelope of each articulated vehicle.

[0012] Optionally, determining the marking scheme based on the trajectory repeatability, the geometric constraints, and the hazard distribution characteristics includes: determining several marking positions based on the trajectory range; determining the marking confidence level for each marking position based on the trajectory repeatability; determining the minimum allowable width based on the geometric constraints; determining the compensation intensity based on the hazard distribution characteristics; and determining the optimal marking position and compensation scheme based on the marking confidence level for each marking position, the minimum allowable width, and the compensation intensity, thereby obtaining the marking scheme.

[0013] Optionally, after determining the optimal marking position and compensation scheme based on the marking confidence level, the minimum allowable width, and the compensation intensity at each marking position, and obtaining the marking scheme, the method further includes: determining a vehicle dynamics constraint set based on the historical trajectory dataset; constructing a driving psychological expectation model based on the historical trajectory dataset; analyzing the coupling relationship between the vehicle dynamics constraint set and the driving psychological expectation model; dynamically compensating and correcting the marking scheme to generate the final marking drawing scheme.

[0014] Secondly, this application provides a road marking drawing system based on image recognition, the system comprising:

[0015] The data acquisition module is used to acquire historical trajectory datasets, road design parameter sets, and environmental image sets for the target road.

[0016] The data analysis module is used to analyze the historical trajectory dataset and determine the trajectory repetition rate;

[0017] The parameter analysis module is used to determine the geometric constraints based on the road design parameter set.

[0018] The image analysis module is used to determine the distribution characteristics of hazard sources based on the environmental image set;

[0019] The scheme determination module is used to determine the marking scheme based on the trajectory repeatability, the geometric constraints, and the hazard source distribution characteristics.

[0020] Thirdly, this application provides an electronic device, including: a memory and a processor;

[0021] The memory is used to store program instructions;

[0022] The processor is configured to call and execute program instructions in the memory to perform the method as described in any of the first aspects. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram illustrating an application scenario provided in one embodiment of this application;

[0025] Figure 2 A flowchart illustrating a road marking drawing method based on image recognition, provided in one embodiment of this application;

[0026] Figure 3 A flowchart illustrating another image recognition-based road marking drawing system provided in an embodiment of this application;

[0027] Figure 4 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0028] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0029] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.

[0030] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.

[0031] The accident-inducing rate of existing road marking schemes is higher than that of ordinary road sections. Therefore, how to solve the significant technical defects of the current road marking drawing method in the application of complex terrain road sections has become a major core issue.

[0032] Based on this, this application provides a road marking drawing method and system based on image recognition. The method involves acquiring historical trajectory datasets, road design parameter sets, and environmental image sets for the target road; analyzing the historical trajectory dataset to determine trajectory repeatability; determining geometric constraints based on the road design parameter set; determining hazard source distribution characteristics based on the environmental image set; and determining the marking scheme based on trajectory repeatability, geometric constraints, and hazard source distribution characteristics. Extracting historical vehicle trajectory datasets helps improve trajectory classification accuracy. Extracting the road design parameter set helps improve the separation of vehicle type-related trajectory features. Extracting the environmental image set helps provide data support for hazard source distribution characteristics, compensating for the shortcomings of traditional static recognition. Analyzing the historical trajectory dataset helps eliminate GPS drift and abnormal driving behavior data, while effectively reducing the extraction error of trajectory range. Calculating trajectory repeatability helps improve the warning effect of markings and reduce the critical danger state of insufficient safety redundancy caused by trajectory competition between small vehicles and articulated vehicles. Generating geometric constraints helps reduce the rate of skid accidents and judgment errors. Determining the distribution characteristics of hazard sources helps improve the visibility distance of road markings under extreme conditions, reduces accidents caused by speeding, and minimizes conflicts in hazard avoidance space on the roadside. Generating road marking schemes helps avoid compensation intensity exceeding lane physical limitations and reduces decision-making error rates at abrupt changes in road markings.

[0033] Figure 1 This application provides an illustration of a scenario where the method provided is applied to road marking in mountainous areas. Specifically, the method is used on any server, which interacts with vehicle-mounted GPS, roadside radar, video surveillance equipment, databases, and drones in the mountainous area to acquire historical trajectory datasets collected by the vehicle-mounted GPS, roadside radar, and video surveillance equipment, road design parameter sets retrieved from the database, and environmental image sets collected by drone aerial photography. The historical trajectory datasets, road design parameter sets, and environmental image sets are carefully analyzed, and a marking scheme is determined based on the analysis results to avoid compensation intensity exceeding lane physical limitations and reduce the decision-making error rate at abrupt changes in marking. Specific implementation details can be found in the following embodiments.

[0034] Figure 2 This is a flowchart illustrating a road marking drawing method based on image recognition, provided as an embodiment of this application. The method of this embodiment can be applied to the server in the above scenario. For example... Figure 2 As shown, the method includes:

[0035] S201. Obtain the historical trajectory dataset, road design parameter set, and environmental image set of the target road;

[0036] The target road can be a mountain road or a complex terrain section that includes continuous curves, steep slopes and dangerous roadside environments.

[0037] Historical trajectory datasets can be collections of vehicle driving trajectory data over a past period. This past period can be selected based on experience or defined subjectively.

[0038] A road design parameter set can be a set of digital parameters of road geometric features.

[0039] The environmental image set can be a multimodal image dataset of the road environment.

[0040] Specifically, historical vehicle trajectory datasets for the target road are collected using data acquisition devices such as vehicle-mounted GPS, roadside radar, and video surveillance equipment. Road design parameters, including curve radii, longitudinal slope, and superelevation, are extracted from road design documents in the database. Multi-view road images are acquired through drone aerial photography. A semantic segmentation model is constructed based on these images. The semantic segmentation model is then used to extract environmental image sets showing roadside cliff boundaries, guardrail locations, and meteorological characteristics.

[0041] S202. Analyze historical trajectory datasets to determine trajectory repetition.

[0042] Trajectory repeatability can be an indicator that quantifies the degree of spatial overlap in vehicle trajectories.

[0043] Specifically, cluster analysis is performed on the historical trajectory dataset to distinguish between articulated vehicle and small vehicle trajectory clusters; then, the turning phase features and straight-line maintenance rates of the articulated vehicle and small vehicle trajectories are extracted. A three-dimensional coordinate system is established based on the road centerline, and timestamps are superimposed to form a spatiotemporal four-dimensional filtering model; a dynamic filtering threshold is set based on the vehicle type, and Kalman filtering is used for trajectory smoothing to remove outliers exceeding the dynamic filtering threshold. Spatial overlay analysis is performed on the filtered historical trajectory dataset to calculate the trajectory repetition at each location point.

[0044] S203. Determine the geometric constraints based on the road design parameter set;

[0045] Geometric constraints can be physical limitations imposed by road alignment on vehicle motion.

[0046] Specifically, the envelope width is corrected by superimposing the superelevation cross slope on the turning radius; the standard deviation of the spatial deviation distribution of the historical trajectory dataset is extracted to calculate the dynamic width requirement. A correlation function between the longitudinal slope and the dynamic width requirement is established. Based on the correlation function, geometric constraints are generated for each road segment.

[0047] S204. Determine the distribution characteristics of hazard sources based on the environmental image set;

[0048] The distribution characteristics of hazardous sources can be quantified parameters of the spatial distribution patterns of hazardous sources.

[0049] Specifically, the distance from the cliff edge to the lane line is obtained through an environmental image set. A digital elevation model is generated using lidar point clouds to calculate the height difference. Radiation fog and advection fog are identified based on image color temperature. A visibility attenuation model is established based on radiation fog and advection fog. The duration of individual fog events and the annual average frequency of fog days are collected through meteorological monitoring. The visibility attenuation model and the duration of individual fog events are calculated to determine the driving impact coefficient. Based on the calculated cliff distance and height difference, the characteristics of high-risk areas are determined. Based on the characteristics of high-risk areas, the driving impact coefficient and the annual average frequency of fog days are calculated to determine the distribution characteristics of hazard sources.

[0050] S205. Determine the marking scheme based on trajectory repeatability, geometric constraints, and hazard source distribution characteristics.

[0051] A road marking scheme can be a standardized implementation plan for traffic markings.

[0052] Specifically, based on the offset between the convex hull vertex of the trajectory range and the road centerline, a candidate location set is generated according to the average distance interval. Trajectory repeatability is used to determine the marking confidence level. Based on geometric constraints, the minimum allowable width is determined by the larger value between the road physical boundary and the dynamic width requirement. The compensation intensity is determined by normalizing the dimensionless safety resource input value calculated based on the hazard source distribution characteristics. Based on the calculated marking confidence level, minimum allowable width, and compensation intensity, an objective function is established to maximize the practicality and adaptability of the markings while satisfying safety and geometric constraints. The marking location and parameter combination that best balances marking confidence level, minimum allowable width, and compensation intensity within the objective function, satisfying safety, geometric constraints, and cost limitations, is selected to generate the final marking scheme.

[0053] This solution improves trajectory classification accuracy by extracting historical vehicle trajectory datasets. Extracting road design parameter sets enhances the separation of vehicle type-related trajectory features. Extracting environmental image sets provides data support for hazard distribution characteristics, compensating for the shortcomings of traditional static identification. Analyzing historical trajectory datasets helps eliminate GPS drift and abnormal driving behavior data, effectively reducing trajectory range extraction errors. Calculating trajectory repeatability improves the warning effect of lane markings and reduces critical danger states caused by insufficient safety redundancy due to trajectory competition between small vehicles and articulated vehicles. Generating geometric constraints helps reduce the rate of skid accidents and judgment errors. Determining hazard distribution characteristics improves the visibility distance of lane markings under extreme conditions, reduces runaway accidents caused by speeding, and reduces conflicts in roadside hazard avoidance space. Generating lane marking schemes helps avoid compensation intensity exceeding lane physical limitations and reduces decision-making error rates at abrupt changes in lane markings.

[0054] In some embodiments, historical trajectory datasets are analyzed to determine the types of vehicles and the corresponding trajectory sets; for each type of vehicle, the trajectory sets are analyzed to determine the spatial deviation distribution of the corresponding vehicle type; a spatiotemporal four-dimensional filtering model is established; based on the spatiotemporal four-dimensional filtering model, abnormal data is removed according to the spatial deviation distribution, and the trajectory range of the removed trajectory sets is determined according to the spatial deviation distribution; the trajectory repeatability is determined according to the trajectory range.

[0055] Vehicle types can be classified based on their physical parameters, such as small vehicles and articulated vehicles.

[0056] The corresponding type can be the vehicle type matched by the classification rules.

[0057] A driving trajectory set can be a collection of historical trajectory data generated by the same type of vehicle on a target road.

[0058] The corresponding vehicle type can be the same as the vehicle type in the trajectory dataset.

[0059] Spatial deviation distribution can be the statistical distribution of the lateral offset of a vehicle trajectory set relative to the road centerline.

[0060] The spatiotemporal four-dimensional filtering model can be a trajectory cleaning model that integrates timestamps, longitude, latitude, and elevation.

[0061] Abnormal data can be trajectory data that does not meet geometric constraints.

[0062] The trajectory range can be the spatial boundary formed by the valid trajectory points retained after filtering.

[0063] Specifically, historical trajectory datasets are accessed to extract vehicle feature parameters such as vehicle length, wheelbase, and maximum steering angle for each trajectory. Based on these parameters, a vehicle type classification rule base is established: vehicles with a larger wheelbase and articulated segment feature signals are classified as articulated vehicles; vehicles with a smaller wheelbase and larger maximum steering angle are classified as small vehicles. The classification results are then associated with the corresponding historical trajectory datasets to generate a set of driving trajectories. A local coordinate system is established using the road centerline as a baseline, and spatiotemporal coordinate normalization is performed on the trajectories of the same vehicle type. Road cross-sections are divided according to average distance intervals, and the lateral offset distribution of trajectory points corresponding to the vehicle type within each cross-section is statistically analyzed, i.e., the spatial deviation distribution. Based on the spatiotemporal data statistical distribution theory, this paper establishes a spatiotemporal four-dimensional filtering model by integrating time series analysis (such as Kalman filtering) and spatial clustering methods (such as DBSCAN), and adjusting the filtering intensity by combining dynamic threshold settings (such as the 3σ criterion) and road segment geometric features (such as curve radius). This model identifies and eliminates abnormal trajectory data. A dynamic threshold is then set according to the spatial deviation distribution. Trajectory points exceeding the dynamic threshold are marked as abnormal data; if consecutive trajectory points show abnormal data, the entire trajectory is eliminated. Furthermore, enhanced filtering is applied to sharp curves, reducing the dynamic threshold to generate the trajectory range. Convex hull analysis is performed on the filtered trajectory range to generate a spatial envelope covering the trajectory points. The curvature continuity of the spatial envelope is optimized by combining the road longitudinal slope and superelevation. Based on the optimized spatial envelope, the critical safe distance between the spatial envelope and the road boundary is calculated. The width of the road centerline is expanded according to the spatial deviation distribution. Finally, the trajectory overlap complexity of each vehicle type within the expanded baseline range is statistically analyzed.

[0064] This solution generates a set of driving trajectories, which helps quantify the differences in driving characteristics for each vehicle type, eliminates the problem of misjudging the trajectory of articulated vehicles, and improves the warning effect of curve markings. Statistical analysis of spatial deviation distribution helps reduce the rate of skidding accidents. Establishing a spatiotemporal four-dimensional filtering model helps filter out abnormal data under adverse weather conditions, solves the problem of trajectory distortion during rush hour and in adverse weather, reduces trajectory point drift caused by visibility decay, shortens the driver's trajectory correction delay, and effectively reduces the rate of speeding and loss of control accidents. Determining the trajectory overlap complexity helps increase the continuity of visual guidance in sections of abrupt changes in markings, reducing the braking response delay.

[0065] In some embodiments, for articulated vehicles, the driving trajectory sets of articulated vehicles and small vehicles are analyzed to obtain a first vehicle analysis result for the articulated vehicles and a second vehicle analysis result for the small vehicles; based on the first vehicle analysis result, the turning envelope of each articulated vehicle is determined; based on the turning envelope, the turning boundary of the articulated vehicle is determined; based on the turning boundary, the trajectory range of the eliminated driving trajectory set is determined according to the second vehicle analysis result and the spatial deviation distribution.

[0066] Articulated vehicles can be multi-axle transport vehicles with an articulated structure.

[0067] The first vehicle analysis result can be the analysis result of the set of articulated vehicle driving trajectories.

[0068] The second vehicle analysis result can be the analysis result of the set of driving trajectories of small vehicles.

[0069] The turning envelope can be a dynamic spatial boundary generated based on the first vehicle analysis results of the articulated vehicle.

[0070] The turning boundary can be the final safety boundary generated by integrating the safety redundancy compensation mechanism, the super-high cross slope, and the overlapping area of ​​the trajectories of articulated vehicles and small vehicles on the basis of the turning envelope.

[0071] Specifically, the maximum lateral offset, turning radius extreme points, and rear axle offset dynamic parameters of the articulated vehicle's trajectory set are extracted; thus, a trajectory curvature variation map of the articulated vehicle is constructed, and the abrupt change characteristics of the towed trajectory at continuous curves are labeled, which is the first vehicle analysis result. The average lateral offset, speed change gradient, and braking point distribution density of the small vehicle's trajectory set are statistically analyzed; based on vehicle kinematics theory (such as the inverse relationship between speed and curvature)... ,in The coefficient of friction, To quantify the trajectory deformation characteristics of small vehicles caused by speed adjustments in curves, a trajectory curvature-speed correlation model is established based on the turning radius and driving behavior statistics (fitting an empirical regression equation of curvature-speed from historical trajectory data). This model quantifies the trajectory compression characteristics at curves, representing the second vehicle analysis result. Based on this second vehicle analysis result, trajectory points with quantile values ​​of lateral offset in the articulated vehicle's trajectory are selected as envelope reference points. The spatial distribution weights of the envelope reference points are dynamically adjusted according to the load status. Several envelope reference points are then connected to generate an initial convex hull boundary. Based on the longitudinal slope, the initial convex hull boundary is longitudinally stretched and corrected. This is then superimposed with the dynamic parameters of the articulated vehicle's rear axle offset to generate the turning envelope for each articulated vehicle. The lateral distance between the turning envelope and the road boundary is calculated. When the distance is too small, a safety redundancy compensation mechanism is triggered. The safety redundancy compensation mechanism is adjusted according to the superelevation cross slope to generate a slope-correlated turning boundary. The turning boundary of the articulated vehicle is superimposed with the analysis results of the second vehicle to generate an initial trajectory range; the initial trajectory range is then subjected to probability-weighted smoothing based on the probability density function of the spatial deviation distribution; thus determining the trajectory range of the final driving trajectory set.

[0072] This solution determines the first vehicle analysis results for articulated vehicles and the second vehicle analysis results for smaller vehicles, providing a dynamic parameter benchmark for generating the turning envelope, reducing the misjudgment rate of towed trajectories at curves, and improving the accuracy of lane marking warning matching. Generating the turning envelope helps avoid the problem of insufficient critical safety distance between the driving trajectories of articulated vehicles and smaller vehicles. Generating the turning boundary helps reduce the rate of sideslip accidents, reduces the decision-making error rate caused by abrupt changes in lane markings, and eliminates trajectory deviations caused by superelevation, improving the continuity of lane markings at curves. Determining the trajectory range helps trigger driver trajectory corrections in advance, reducing the rate of loss of control accidents caused by speeding.

[0073] In some embodiments, the road design parameter set is analyzed to obtain the curve radius, longitudinal slope, and superelevation; the spatial deviation distribution of the turning envelope and small vehicles is analyzed to determine the dynamic width requirement; the dynamic width requirement is corrected according to the coupling relationship between the longitudinal slope and the superelevation to obtain the actual dynamic width requirement; and the actual dynamic width requirement is determined as a geometric constraint condition.

[0074] The radius of a curve can be the radius of curvature of the road centerline in the horizontal plane.

[0075] Longitudinal slope can be the longitudinal slope value of a road.

[0076] Super-high cross slope can be the lateral slope of the road surface at a curve.

[0077] The dynamic width requirement can be the initial value of the minimum lane width.

[0078] The coupling relationship can be the interaction mechanism between the longitudinal slope gradient and the superelevation cross slope.

[0079] The actual dynamic width requirement can be the final lane width value after correction.

[0080] Specifically, the original road design parameter set is obtained, and three core geometric parameters—curving radius, longitudinal slope, and superelevation cross slope—are extracted. Based on the turning envelope of articulated vehicles, the maximum lateral width of the turning envelope is calculated. Spatial clustering is performed on the driving trajectory set of small vehicles, and driving trajectory points with spatial deviation distribution are selected to calculate the width of the trajectory dense area for small vehicles. Combining the maximum lateral width, the width of the trajectory dense area, and the safety redundancy coefficient, the dynamic width requirement is calculated. A compensation coefficient formula for the coupling relationship between longitudinal slope and superelevation cross slope is established. When the compensation coefficient is too high, the slope-related compensation mode is triggered. The dynamic width requirement is multiplied by the compensation coefficient to generate the corrected dynamic width requirement. An additional longitudinal stretching compensation is added for downhill sections. The lateral distance between the corrected dynamic width requirement and the road boundary is verified. If the lateral distance is insufficient, the boundary is expanded outward according to the difference to generate the final actual dynamic width requirement. The actual dynamic width requirement is bound to the curve radius, longitudinal slope, and superelevation, forming a dynamic geometric constraint tuple that includes the curve radius, longitudinal slope, superelevation, and actual width. According to road design specifications, the actual width is converted into lane marking design parameters: lane spacing is calculated based on the actual width; simultaneously, edge marking offset is calculated based on the actual width and safety redundancy factor. Geometric constraints are then generated based on the lane spacing and edge marking offset.

[0081] This scheme extracts curve radius, longitudinal slope, and superelevation cross slope to provide basic variables for dynamic correction. Based on the turning envelope of articulated vehicles, the maximum lateral width of the turning envelope is calculated to ensure that the marking design fully covers the maximum lateral offset of articulated vehicles, reducing the problem of drivers misjudging towing trajectories. The width of the trajectory density area for small vehicles is calculated to identify trajectory overlap risk areas and correct defects caused by insufficient safety redundancy space. Calculating dynamic width requirements helps to achieve dynamic compensation of the side slip accident rate on downhill sections. Generating the final actual dynamic width requirements helps to reduce side slip accidents caused by fixed lane widths, offset the vehicle center of gravity shift effect caused by superelevation cross slopes, and prevent vehicles from crossing the line due to centrifugal force. Generating geometric constraints helps to reduce the probability of marking warning failure and also helps the marking scheme adapt to road geometry changes in real time.

[0082] In some embodiments, an environmental image set is analyzed to determine road and meteorological information; based on the road information, the distance and height difference between the roadside cliffs are determined; based on the meteorological information, the annual average frequency of fog days and the duration of a single fog event are determined; based on the meteorological information, the fog type is determined; based on the fog type and meteorological information, the driving impact of each type of fog is determined; based on the distance and height difference between the roadside cliffs, the characteristics of high-risk areas are determined; based on the annual average frequency of fog days, driving impact, and duration of a single fog event, visibility hazard characteristics are determined; and the characteristics of high-risk areas and visibility hazard characteristics are identified as hazard source distribution characteristics.

[0083] Road information can be obtained by parsing road geometric feature data from environmental image sets.

[0084] Meteorological information can be meteorological state parameters such as fog concentration level, precipitation probability, and wind speed extracted from environmental image sets, meteorological sensor data, and image time series information.

[0085] The distance to the roadside cliff can be the shortest horizontal distance from the road edge line to the nearest cliff or valley edge.

[0086] The height difference can be the vertical elevation difference between the road surface and the bottom of a cliff or dangerous terrain.

[0087] The annual average frequency of foggy days can be the cumulative number of foggy days in a target area within a year.

[0088] The duration of a single fog event can be the continuous time from the formation of the fog event to its dissipation.

[0089] Fog type can be a classification result of fog.

[0090] Fog characteristics can be a combination of physical properties of fog types.

[0091] Driving impact can be a quantitative indicator of how fog reduces a driver's perception and control abilities.

[0092] High-risk area characteristics can be a composite risk characterization parameter composed of the distance and height difference between the roadside cliffs.

[0093] Visibility hazard characteristics can be a comprehensive meteorological risk indicator that integrates the annual average frequency of fog days, the duration of a single fog event, and the impact on driving.

[0094] Specifically, pixel-level fusion is performed on the environmental image set, and road information such as road surface texture, shoulder contour, and slope morphology features are extracted using a convolutional neural network. Simultaneously, meteorological sensor data is matched with the image time-series information to determine meteorological information. A depth map is generated based on binocular camera images, and the horizontal projection length from the road edge to the cliff boundary is calculated, i.e., the distance to the roadside cliff. Point cloud elevation difference analysis is used to determine the elevation difference between the road surface and the bottom of the cliff, i.e., the height difference. Spatiotemporal labels of the environmental image set are correlated with a meteorological database to statistically analyze the annual average frequency of fog days and the duration of individual fog events. Based on meteorological information, fog types are classified according to their formation mechanisms: radiation fog is identified by combining image features of diurnal temperature difference and wind speed; advection fog is identified by combining humidity gradient and wind direction persistence; and uphill fog is identified by topographic elevation difference and humidity diffusion patterns. Fog concentration gradient values ​​from the environmental image set are extracted through image chromaticity analysis and matched with measured visibility values ​​obtained from meteorological sensors to predefined fog type classification thresholds. Based on the matching results, the light scattering coefficient and humidity saturation of the corresponding fog type are retrieved from a historical fog database. Based on the light scattering coefficient of the fog type, a classification method is adopted... The system calculates dynamic visibility distance using a law; based on the dynamic visibility distance, a piecewise function model is used to calculate the trajectory correction error rate; based on the trajectory correction error rate, standard visibility distance, and dynamic visibility distance, a driving impact coefficient is generated using weighted fusion. A high-risk threshold is defined based on the distance and height difference between the roadside cliff and the roadside area; the boundary of the danger zone is delineated based on the threshold, generating high-risk area characteristics. A visibility hazard index is calculated based on the annual average fog frequency, driving impact coefficient, and duration of a single fog event; a visibility-marking correction relationship is established based on the visibility hazard index; and visibility hazard characteristics are determined by combining the visibility hazard index and the visibility-marking correction relationship. High-risk area characteristics are spatiotemporally matched with visibility hazard characteristics; based on the matching results, road sections that simultaneously meet the criteria of high cliff risk and high visibility risk are marked as composite hazard sources, and a dynamic marking enhancement mode is activated; thus, a hazard source distribution characteristic containing four-dimensional parameters—hazard type, spatial location, duration, and risk level—is generated.

[0095] This solution extracts road information, which helps improve the accuracy of road boundary identification; it also determines meteorological information, providing spatiotemporal matching data support for hazard source quantification; determining elevation differences helps improve the positioning accuracy of high-risk areas; statistically analyzing the annual average frequency of fog days and the duration of individual fog events helps avoid the problem of visibility attenuation; and identifying fog types helps reduce the decision-making error rate and avoid the misjudgment rate of mixed fog. Calculating dynamic visibility distance using laws helps reduce driver prediction errors. Determining the driving impact coefficient for each type of fog helps reduce the rate of speeding and loss of control accidents. Generating high-risk area features helps reduce rollover accidents and the safety redundancy issues caused by trajectory competition between articulated vehicles and small vehicles. Identifying visibility hazard features helps reduce driver speed perception distortion. Generating hazard source distribution features helps shorten the dynamic adjustment response time of lane marking schemes, thus reducing the driver's decision-making error rate.

[0096] In some embodiments, the rear axle offset data of the articulated vehicle is determined based on the analysis results of the first vehicle; the vehicle load status is predicted based on the rear axle offset data; the dynamic envelope range is determined based on the vehicle load status and the rear axle offset data; the dynamic envelope range is corrected based on the coupling relationship between the rear axle offset data and the superelevation cross slope, thereby obtaining the turning envelope of each articulated vehicle.

[0097] Rear axle offset data can be the lateral displacement of the rear axle centerline relative to the theoretical driving trajectory during the driving process of an articulated vehicle.

[0098] The vehicle load status can be the vehicle load level.

[0099] The dynamic envelope range can be the boundary of the vehicle's turning space requirement.

[0100] Specifically, real-time rear axle offset is acquired using onboard displacement sensors and combined with the baseline wheelbase from the first vehicle analysis results to calculate rear axle offset data. Based on the rear axle offset data, load levels are classified, and a random forest algorithm is used to fuse the number of axle load sensors and the rear axle offset data to predict the vehicle's load status. Based on the rear axle offset data corresponding to the vehicle's load status and the first vehicle analysis results, the minimum turning radius is calculated. Using the minimum turning radius as a benchmark, a safety redundancy coefficient is added to generate a dynamic envelope range. Based on the minimum turning radius, the equivalent offset increment caused by the superelevation is calculated; based on the equivalent offset increment, the lateral expansion of the envelope is corrected; and by integrating the dynamic envelope range and the lateral expansion of the envelope, a turning envelope for each articulated vehicle is generated.

[0101] This scheme calculates rear axle offset data, providing a data foundation for calculating the dynamic envelope range. Predicting vehicle load status helps reduce the alignment error of the lane markings in the trajectory competition area between articulated vehicles and smaller vehicles. Generating the dynamic envelope range helps reduce the rate of sideslip accidents and the error in predicting corner radii, shortening the driver's braking response delay. Generating the turning envelope for each articulated vehicle helps reduce the decision-making error rate.

[0102] In some embodiments, several marking positions are determined based on the trajectory range; the marking confidence level for each marking position is determined based on the trajectory repeatability, using the following formula:

[0103] (1)

[0104] in, Indicates the confidence level of the gradation line; ; ; Represents the discrete variance of the trajectory;

[0105] The minimum allowable width is determined based on geometric constraints using the following formula:

[0106] (2)

[0107] in, Indicates the minimum allowable width; ; The minimum permissible width for a single lane as specified in the national road design standards is 3.5 meters;

[0108] Based on the distribution characteristics of the hazard sources, the compensation intensity is determined; based on the confidence level of the marking, the minimum allowable width, and the compensation intensity at each marking location, the optimal marking location and compensation scheme are determined, resulting in the marking scheme.

[0109] The location of the marking can be a set of candidate marking placement points.

[0110] Marking confidence can be an indicator that quantifies the degree of matching between the locations of several markings and the actual driving paths of vehicles in historical trajectory data.

[0111] The minimum allowable width can be the minimum physical width of the lane.

[0112] The compensation intensity can be the level of adjustment of the marking parameters set for different types of hazards.

[0113] The optimal location for road markings can be the best point for laying out the road markings.

[0114] The compensation scheme can be a set of strategies for adjusting the datum parameters.

[0115] Specifically, based on the spatial distribution characteristics of the trajectory range, a density clustering algorithm is used to identify high-frequency trajectory areas and eliminate discrete noise points. Using the geometric center of the dense trajectory area as a reference, several lane marking locations are generated using a dynamic interval algorithm. A trajectory repeatability index is defined, and the trajectory repeatability of several lane marking locations in historical trajectory data is statistically analyzed to select the lane marking confidence level for each location. Based on the lateral force balance equation under the coupled action of longitudinal slope and superelevation, a slope correction coefficient is calculated, and the lane width benchmark value is dynamically adjusted, thereby deriving the minimum allowable width. According to the hazard source distribution characteristics, a risk-visibility attenuation mapping table is constructed, and compensation levels are assigned according to hazard source type. Based on real-time meteorological data, the compensation intensity is determined according to the compensation level. Based on the lane marking confidence level, minimum allowable width, and compensation intensity for each lane marking location, a multi-objective optimization model is established by integrating multi-objective optimization theory, traffic engineering specifications, driving behavior models, and empirical data. Pareto optimal solutions are then used to select the final lane marking scheme. (Track coverage frequency...) This indicates the percentage of a lane marking location covered by historical tracks, reflecting the driver's actual route selection preferences. This indicates the number of trajectory points that fall within the buffer zone of the candidate location for the caliper (e.g., ±0.5m). This represents the total number of trajectory points for this road segment; the trajectory discrete variance penalty term. This indicates a correction for the decrease in confidence caused by the dispersion of trajectory distribution. Variance The larger the value, the more significant the confidence decay. If the trajectory is highly concentrated ( The penalty term approaches 1, and the confidence level is dominated by the coverage frequency. If the trajectory is discrete ( As the penalty term approaches zero, the confidence level decreases. (Exponential function) It can quickly attenuate the effects of high variance, which aligns with the common understanding in driving behavior that "the higher the trajectory consistency, the stronger the effectiveness of the lane markings"; among which The specific settings need to be determined through multiple trials. The existing "Technical Standards for Highway Engineering" Article 4.0.2 stipulates that when a Class IV highway uses a single lane, the lane width should be 3.5 meters; when dynamic demands... The results are based on dynamic calculations and are adapted to complex terrain; if the dynamic calculation values ​​are affected by insufficient data or errors... A mandatory 3.5m measurement was adopted to ensure legal compliance.

[0116] This scheme eliminates discrete noise points, helping to reduce interference trajectories caused by driver error or temporary obstacle avoidance; generates several marking locations, helping to reduce trajectory conflict rates for each vehicle type. Trajectory repeatability helps to identify frequently used trajectory areas in actual driving. Determining marking confidence levels helps to reduce the impact of abnormal trajectories such as speeding or low-speed driving on confidence levels, reducing the misjudgment rate of curve trajectories. Determining the minimum permissible width helps to improve the width adaptation of markings on downhill sections, reducing the rate of skidding accidents and the incidence of critical dangerous states. Determining the compensation intensity helps to shorten the driver's trajectory correction delay time, increase the safe distance between the vehicle and roadside hazards, and reduce the driver's prediction error of curve radii. Generating the final marking scheme helps to reduce the error rate of decision-making at abrupt changes in markings and improves the timeliness of risk response.

[0117] In some embodiments, a vehicle dynamics constraint set is determined based on a historical trajectory dataset; a driving psychological expectation model is constructed based on the historical trajectory dataset; the coupling relationship between the vehicle dynamics constraint set and the driving psychological expectation model is analyzed, and the lane marking scheme is dynamically compensated and corrected to generate the final lane marking drawing scheme.

[0118] The vehicle dynamics constraint set can be a set of dynamic parameters that include the type of vehicle being driven and its load status.

[0119] A driving psychological expectation model can be a quantitative model constructed through the analysis of driver cognitive characteristics and behavioral data.

[0120] The final line marking scheme can be a line marking parameter instruction set generated after data fusion and correction.

[0121] Specifically, trajectory samples of articulated vehicles and small vehicles are separated from historical trajectory datasets. Then, based on vehicle wheelbase, track width, and maximum steering angle, the turning envelope range of articulated vehicles is calculated, and the spatial competition relationship between the trajectories of articulated vehicles and small vehicles is measured. Combining the curve radius and longitudinal slope, the lateral offset threshold of the vehicle is derived, thereby determining the vehicle dynamics constraint set. The driver's prediction error regarding the curve radius and braking response delay are statistically analyzed to generate a prediction error threshold. A correlation mapping is established between the speed perception distortion rate on longitudinal slope sections and road marking warnings to generate a reaction time window. The correlation between the decision error rate at abrupt changes in road markings and the interval between markings is analyzed, and the driver's success rate in correcting the trajectory of gradually changing road markings is extracted to generate a visual continuity requirement. Through cognitive psychology, human factors engineering, control theory, and empirical driving research, a multidisciplinary approach is used to dynamically couple the prediction error threshold, reaction time window, and visual continuity requirement into a driving psychological expectation model. The system compares the vehicle dynamics constraint set with the driver's psychological expectation model in real time. If a conflict continues, it initiates dynamic adjustment of the marking position. Then, it iteratively solves the marking parameters corresponding to the maximum value of the objective function using the gradient descent method. The marking parameters are imported into the simulation environment, and road design parameters are set. If the trajectory deviation exceeds the standard, the node coordinates are called back until the standard is met. Vehicle trajectory data is collected in real time, and the latest confidence level is calculated. Finally, the sliding window algorithm is used to smooth the weight fluctuations and generate the final marking drawing scheme.

[0122] This solution separates the trajectory samples of articulated vehicles and small vehicles, helping to reduce the misjudgment rate of vehicle trajectories and improve the warning effect of road markings. Determining the vehicle dynamics constraint set helps reduce the rate of skidding accidents and accidents on high-risk road sections, increasing safety redundancy. Determining the driver's psychological expectation model helps drivers shorten braking response delay time, reducing the accident rate on curves, while also reducing speed perception distortion and minimizing speeding and loss-of-control accidents. Generating the final road marking drawing scheme contributes to a long-term stable rate of accident reduction.

[0123] Figure 3 A schematic diagram of a road marking drawing system based on image recognition is provided in one embodiment of this application, as shown below. Figure 3 As shown, the road marking drawing system 300 based on image recognition in this embodiment includes: a data acquisition module 301, a data analysis module 302, a parameter analysis module 303, an image analysis module 304, and a scheme determination module 305.

[0124] Data acquisition module 301 is used to acquire historical trajectory datasets, road design parameter sets, and environmental image sets of the target road;

[0125] Data analysis module 302 is used to analyze the historical trajectory dataset and determine the trajectory repetition rate;

[0126] The parameter analysis module 303 is used to determine the geometric constraints based on the road design parameter set;

[0127] Image analysis module 304 is used to determine the distribution characteristics of hazard sources based on the environmental image set;

[0128] The scheme determination module 305 is used to determine the marking scheme based on the trajectory repeatability, the geometric constraints, and the hazard source distribution characteristics.

[0129] Optionally, when the data analysis module 302 analyzes the historical trajectory dataset and determines the trajectory repetition, it is used for:

[0130] Analyze the historical trajectory dataset to determine the types of vehicles and the corresponding trajectory sets.

[0131] For each type of vehicle, the set of driving trajectories is analyzed to determine the spatial deviation distribution of the corresponding vehicle type.

[0132] A spatiotemporal four-dimensional filtering model is established. Based on the spatiotemporal four-dimensional filtering model, abnormal data is removed according to the spatial deviation distribution. The trajectory range of the removed driving trajectory set is determined according to the spatial deviation distribution.

[0133] Based on the trajectory range, the trajectory repeatability is determined.

[0134] Optionally, the vehicle type includes articulated vehicles and small vehicles. When the data analysis module 302 determines the trajectory range of the eliminated driving trajectory set based on the spatial deviation distribution, it is used for:

[0135] For the articulated vehicle, the driving trajectory set of the articulated vehicle and the driving trajectory set of the small vehicle are analyzed to obtain the first vehicle analysis result of the articulated vehicle and the second vehicle analysis result of the small vehicle.

[0136] Based on the analysis results of the first vehicle, the turning envelope of each articulated vehicle is determined;

[0137] The turning boundary of the articulated vehicle is determined based on the turning envelope.

[0138] Based on the turning boundary, and according to the second vehicle analysis results and the spatial deviation distribution, the trajectory range of the eliminated driving trajectory set is determined.

[0139] Optionally, when the parameter analysis module 303 determines the geometric constraints based on the road design parameter set, it is used for:

[0140] The road design parameter set is analyzed to obtain the curve radius, longitudinal slope, and superelevation cross slope.

[0141] Analyze the turning envelope and the spatial deviation distribution of small vehicles to determine the dynamic width requirement;

[0142] Based on the coupling relationship between the longitudinal slope and the superelevation cross slope, the dynamic width requirement is corrected to obtain the actual dynamic width requirement;

[0143] The actual dynamic width requirement is determined as a geometric constraint.

[0144] Optionally, when the image analysis module 304 determines the distribution characteristics of the hazard source based on the environmental image set, it is used to:

[0145] Analyze the environmental image set to determine road information and meteorological information;

[0146] Based on the road information, determine the distance and height difference between the roadside cliffs;

[0147] Based on the meteorological information, the annual average frequency of foggy days and the duration of a single fog event are determined;

[0148] Based on the meteorological information, determine the type of fog;

[0149] Based on the fog type and the meteorological information, determine the driving impact of each type of fog;

[0150] Based on the distance and height difference between the roadside cliffs, the characteristics of high-risk areas are determined;

[0151] Based on the annual average frequency of foggy days, the driving impact, and the duration of a single fog event, visibility hazard characteristics are determined;

[0152] The high-risk area characteristics and the visibility hazard characteristics are identified as hazard source distribution characteristics.

[0153] Optionally, when the data analysis module 302 determines the turning envelope of each articulated vehicle based on the analysis results of the first vehicle, it is used to:

[0154] Based on the analysis results of the first vehicle, the rear axle offset data of the articulated vehicle is determined;

[0155] Based on the rear axle offset data, predict the vehicle load status;

[0156] The dynamic envelope range is determined based on the vehicle load status and the rear axle offset data.

[0157] Based on the rear axle offset data and the coupling relationship of the superelevation cross slope, the range of the dynamic envelope is corrected to obtain the turning envelope of each articulated vehicle.

[0158] Optionally, when the scheme determination module 305 determines the marking scheme based on the trajectory repeatability, the geometric constraints, and the hazard source distribution characteristics, it is used to:

[0159] Based on the trajectory range, determine the positions of several markings;

[0160] Based on the trajectory repeatability, determine the confidence level of the marking position for each marking.

[0161] Determine the minimum allowable width based on the aforementioned geometric constraints;

[0162] The compensation intensity is determined based on the distribution characteristics of the hazard sources.

[0163] Based on the confidence level of the marking at each location, the minimum allowable width, and the compensation intensity, the optimal marking location and compensation scheme are determined, resulting in a marking scheme.

[0164] Optionally, the image recognition-based road marking drawing system further includes a scheme generation module 306, used for:

[0165] Based on the historical trajectory dataset, determine the set of vehicle dynamics constraints;

[0166] Based on the historical trajectory dataset, a driving psychological expectation model is constructed;

[0167] The coupling relationship between the vehicle dynamics constraint set and the driver's psychological expectation model is analyzed, and the marking scheme is dynamically compensated and corrected to generate the final marking scheme.

[0168] The system in this embodiment can be used to execute the methods of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

[0169] Figure 4 This is a schematic diagram of the structure of an electronic device provided in one embodiment of this application, such as... Figure 4 As shown, the electronic device 400 of this embodiment may include a memory 401 and a processor 402.

[0170] The memory 401 stores a computer program that can be loaded by the processor 402 and execute the methods described in the above embodiments.

[0171] The processor 402 and the memory 401 are connected, for example, via a bus.

[0172] Optionally, the electronic device 400 may also include a transceiver. It should be noted that in practical applications, the transceiver is not limited to one, and the structure of the electronic device 400 does not constitute a limitation on the embodiments of this application.

[0173] Processor 402 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It may implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. Processor 402 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0174] A bus can include a pathway for transmitting information between the aforementioned components. The bus can be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. Buses can be categorized as address buses, data buses, control buses, etc. For ease of illustration, only one thick line is used in the diagram, but this does not imply that there is only one bus or one type of bus.

[0175] The memory 401 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto.

[0176] The memory 401 is used to store application code that executes the solution of this application, and its execution is controlled by the processor 402. The processor 402 is used to execute the application code stored in the memory 401 to implement the content shown in the foregoing method embodiments.

[0177] Electronic devices include, but are not limited to: mobile terminals such as mobile phones, laptops, digital radio receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), and in-vehicle terminals (such as in-vehicle navigation terminals), as well as fixed terminals such as digital TVs and desktop computers. Servers can also be included. Figure 4 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0178] The electronic device in this embodiment can be used to execute the method of any of the above embodiments, and its implementation principle and technical effect are similar, so they will not be described again here.

[0179] This application also provides a computer-readable storage medium storing a computer program that can be loaded by a processor and execute the methods described in the above embodiments.

[0180] Those skilled in the art will understand that all or part of the steps of the above-described method embodiments can be implemented by hardware related to program instructions. The aforementioned program can be stored in a computer-readable storage medium. When executed, the program performs the steps of the above-described method embodiments; and the aforementioned storage medium includes various media capable of storing program code, such as ROM, RAM, magnetic disks, or optical disks.

Claims

1. A method for drawing road markings based on image recognition, characterized in that, include: Obtain the historical trajectory dataset, road design parameter set, and environmental image set of the target road; Analyze the historical trajectory dataset to determine the trajectory repetition rate; Determine the geometric constraints based on the road design parameter set; Based on the environmental image set, determine the distribution characteristics of the hazard sources; The marking scheme is determined based on the trajectory repeatability, the geometric constraints, and the hazard source distribution characteristics; The analysis of the historical trajectory dataset to determine trajectory repetition includes: Analyze the historical trajectory dataset to determine the types of vehicles and the corresponding trajectory sets. For each type of vehicle, the set of driving trajectories is analyzed to determine the spatial deviation distribution of the corresponding vehicle type. A spatiotemporal four-dimensional filtering model is established. Based on the spatiotemporal four-dimensional filtering model, abnormal data is removed according to the spatial deviation distribution. The trajectory range of the removed driving trajectory set is determined according to the spatial deviation distribution. Based on the trajectory range, determine the trajectory repeatability; The types of vehicles include articulated vehicles and small vehicles. Determining the trajectory range of the excluded trajectory set based on the spatial deviation distribution includes: By analyzing the trajectory set of the articulated vehicle and the trajectory set of the small vehicle, the first vehicle analysis result of the articulated vehicle and the second vehicle analysis result of the small vehicle are obtained. Based on the analysis results of the first vehicle, the turning envelope of each articulated vehicle is determined; The turning boundary of the articulated vehicle is determined based on the turning envelope. Based on the turning boundary, and according to the second vehicle analysis results and the spatial deviation distribution, the trajectory range of the eliminated driving trajectory set is determined; The step of determining the geometric constraints based on the road design parameter set includes: The road design parameter set is analyzed to obtain the curve radius, longitudinal slope, and superelevation cross slope. Analyze the turning envelope and the spatial deviation distribution of small vehicles to determine the dynamic width requirement; Based on the coupling relationship between the longitudinal slope and the superelevation cross slope, the dynamic width requirement is corrected to obtain the actual dynamic width requirement; The actual dynamic width requirement is determined as a geometric constraint.

2. The method according to claim 1, characterized in that, The step of determining the distribution characteristics of hazard sources based on the environmental image set includes: Analyze the environmental image set to determine road information and meteorological information; Based on the road information, determine the distance and height difference between the roadside cliffs; Based on the meteorological information, the annual average frequency of foggy days and the duration of a single fog event are determined; Based on the meteorological information, determine the fog type; Based on the fog type and the meteorological information, determine the driving impact of each type of fog; Based on the distance and height difference between the roadside cliffs, the characteristics of high-risk areas are determined; Based on the annual average frequency of foggy days, the driving impact, and the duration of a single fog event, visibility hazard characteristics are determined; The high-risk area characteristics and the visibility hazard characteristics are identified as hazard source distribution characteristics.

3. The method according to claim 2, characterized in that, The step of determining the turning envelope of each articulated vehicle based on the analysis results of the first vehicle includes: Based on the analysis results of the first vehicle, the rear axle offset data of the articulated vehicle is determined; Based on the rear axle offset data, predict the vehicle load status; The dynamic envelope range is determined based on the vehicle load status and the rear axle offset data. Based on the rear axle offset data and the coupling relationship of the superelevation cross slope, the range of the dynamic envelope is corrected to obtain the turning envelope of each articulated vehicle.

4. The method according to claim 1, characterized in that, The step of determining the marking scheme based on the trajectory repeatability, the geometric constraints, and the hazard source distribution characteristics includes: Based on the trajectory range, determine the positions of several markings; Based on the trajectory repeatability, determine the confidence level of the marking position for each marking. Determine the minimum allowable width based on the aforementioned geometric constraints; The compensation intensity is determined based on the distribution characteristics of the hazard sources. Based on the confidence level of the marking at each location, the minimum allowable width, and the compensation intensity, the optimal marking location and compensation scheme are determined, resulting in a marking scheme.

5. The method according to claim 4, characterized in that, After determining the optimal marking position and compensation scheme based on the marking confidence level, the minimum allowable width, and the compensation intensity for each marking position, and obtaining the marking scheme, the method further includes: Based on the historical trajectory dataset, determine the set of vehicle dynamics constraints; Based on the historical trajectory dataset, a driving psychological expectation model is constructed; The coupling relationship between the vehicle dynamics constraint set and the driver's psychological expectation model is analyzed, and the marking scheme is dynamically compensated and corrected to generate the final marking scheme.

6. A road marking drawing system based on image recognition, applied to the method described in any one of claims 1-5, characterized in that, include: The data acquisition module is used to acquire historical trajectory datasets, road design parameter sets, and environmental image sets for the target road. The data analysis module is used to analyze the historical trajectory dataset and determine the trajectory repetition rate; The parameter analysis module is used to determine the geometric constraints based on the road design parameter set. The image analysis module is used to determine the distribution characteristics of hazard sources based on the environmental image set; The scheme determination module is used to determine the marking scheme based on the trajectory repeatability, the geometric constraints, and the hazard source distribution characteristics; The analysis of the historical trajectory dataset to determine trajectory repetition includes: Analyze the historical trajectory dataset to determine the types of vehicles and the corresponding trajectory sets. For each type of vehicle, the set of driving trajectories is analyzed to determine the spatial deviation distribution of the corresponding vehicle type. A spatiotemporal four-dimensional filtering model is established. Based on the spatiotemporal four-dimensional filtering model, abnormal data is removed according to the spatial deviation distribution. The trajectory range of the removed driving trajectory set is determined according to the spatial deviation distribution. Based on the trajectory range, determine the trajectory repeatability; The types of vehicles include articulated vehicles and small vehicles. Determining the trajectory range of the excluded trajectory set based on the spatial deviation distribution includes: By analyzing the trajectory set of the articulated vehicle and the trajectory set of the small vehicle, the first vehicle analysis result of the articulated vehicle and the second vehicle analysis result of the small vehicle are obtained. Based on the analysis results of the first vehicle, the turning envelope of each articulated vehicle is determined; The turning boundary of the articulated vehicle is determined based on the turning envelope. Based on the turning boundary, and according to the second vehicle analysis results and the spatial deviation distribution, the trajectory range of the eliminated driving trajectory set is determined; The step of determining the geometric constraints based on the road design parameter set includes: The road design parameter set is analyzed to obtain the curve radius, longitudinal slope, and superelevation cross slope. Analyze the turning envelope and the spatial deviation distribution of small vehicles to determine the dynamic width requirement; Based on the coupling relationship between the longitudinal slope and the superelevation cross slope, the dynamic width requirement is corrected to obtain the actual dynamic width requirement; The actual dynamic width requirement is determined as a geometric constraint.

7. An electronic device, characterized in that, include: Memory and processor; The memory is used to store program instructions; The processor is configured to call and execute program instructions in the memory to perform the method as described in any one of claims 1-5.

Images