Parallel Road Detection Method, Device, and Electronic Device

Abstract

The present disclosure provides a parallel road detection method, apparatus, and electronic device, relating to the field of map technology, and particularly to the field of data mining. The specific implementation solution is as follows: obtain a geometric representation line corresponding to the passing direction of a target main road and a target trajectory, where the target trajectory is a trajectory passing through the geometric representation line; generate a trajectory grid map according to the geometric representation line and the target trajectory, and each grid in the trajectory grid map is marked with the number of target trajectories passing through; select multiple detection points from the geometric representation line and generate multiple detection lines corresponding to the multiple detection points; map the multiple detection lines into the trajectory grid map to obtain a target grid map, and extract pixel features of the multiple detection lines from the target grid map, where the pixel features of the multiple detection lines are respectively characterized by the number of target trajectories passing through the grids on the corresponding detection lines; determine a target detection result based on the pixel features of the multiple detection lines, and the target detection result is used to characterize whether there is a parallel road of the target main road.

Description

Technical Field

[0001] The present disclosure relates to the field of map technology, in particular to the field of data mining, and specifically relates to a method, apparatus, and electronic device for parallel road detection. Background Art

[0002] In the related art, one method is to use a collection vehicle to conduct full-coverage collection of a city, and then extract and map according to the trajectory data of the collection vehicle to realize the excavation of parallel roads. However, this method has problems such as high collection cost, low efficiency, and inability to timely supplement newly added parallel roads. Another method is for users to report newly added parallel roads, but the above method has problems such as low recall rate and inability to be scaled up. Summary of the Invention

[0003] The present disclosure provides a method, apparatus, device, and storage medium for parallel road detection.

[0004] According to one aspect of the present disclosure, a method for parallel road detection is provided, including: obtaining a geometric representation line corresponding to the passing direction of a target main road and a target trajectory, where the target trajectory is a trajectory passing through the geometric representation line; generating a trajectory grid map according to the geometric representation line and the target trajectory, where each grid in the trajectory grid map is marked with the number of target trajectories passing through it; selecting a plurality of detection points from the geometric representation line and generating a plurality of detection lines corresponding to the plurality of detection points; mapping the plurality of detection lines into the trajectory grid map to obtain a target grid map, and extracting pixel features of the plurality of detection lines from the target grid map, where the pixel features of the plurality of detection lines are respectively characterized by the number of target trajectories passing through the grids on the corresponding detection lines; determining a target detection result based on the pixel features of the plurality of detection lines, where the target detection result is used to indicate whether there is a parallel road of the target main road.

[0005] Optionally, the generating a trajectory grid map according to the geometric representation line and the target trajectory includes: generating an initial grid map with a predetermined width using the geometric representation line as the center line; respectively obtaining the number of target trajectories passing through each grid in the initial grid map; marking the number of target trajectories passing through each grid in the initial grid map into the corresponding grid to obtain the trajectory grid map.

[0006] Optionally, the selecting a plurality of detection points from the geometric representation line and generating a plurality of detection lines corresponding to the plurality of detection points includes: selecting a plurality of detection points on the geometric representation line based on a predetermined interval; respectively generating a plurality of detection lines with a predetermined length perpendicular to the geometric representation line and corresponding to the plurality of detection points with the plurality of detection points as the centers.

[0007] Optionally, determining the target detection result based on the pixel features of the multiple detection lines includes: respectively determining the pixel main peaks and pixel secondary peaks of the multiple detection lines based on the pixel features of the multiple detection lines; determining the parallel road feature results of the detection points corresponding to the multiple detection lines respectively based on the pixel main peaks and pixel secondary peaks of the multiple detection lines, where the parallel road feature results are used to identify whether the corresponding detection points have parallel road features; and determining the target detection result based on the parallel road feature results of the detection points corresponding to the multiple detection lines.

[0008] Optionally, respectively determining the pixel main peaks and pixel secondary peaks of the multiple detection lines based on the pixel features of the multiple detection lines includes: respectively determining the pixel features of the detection points corresponding to the multiple detection lines as the pixel main peaks of the corresponding detection lines; respectively clustering the pixel features on the right side of the corresponding pixel main peaks on the multiple detection lines to obtain multiple candidate secondary peaks of the corresponding detection lines; and respectively selecting the candidate secondary peak with the largest pixel feature from the multiple candidate secondary peaks of the corresponding detection lines as the pixel secondary peak.

[0009] Optionally, respectively determining the parallel road feature results of the detection points corresponding to the multiple detection lines based on the pixel main peaks and pixel secondary peaks of the multiple detection lines includes: numbering the multiple detection points selected on the geometric representation line in ascending order; respectively clustering the pixel features on the right side of the corresponding pixel main peaks on the multiple detection lines to obtain multiple candidate secondary peaks of the corresponding detection lines, and numbering the multiple candidate secondary peaks in ascending order starting from the detection points; for the target detection point corresponding to the target detection line among the multiple detection lines, obtaining the first serial number of the pixel main peak on the target detection line, and the second serial number of the pixel main peak on the previous detection line of the target detection line in the traffic direction; obtaining the third serial number of the pixel secondary peak on the target detection line, and the fourth serial number of the pixel secondary peak on the previous detection line of the target detection line in the traffic direction; detecting whether the difference between the first serial number and the second serial number is less than a first difference threshold, and detecting whether the difference between the third serial number and the fourth serial number is less than a second difference threshold; and determining that the target detection point has parallel road features when the detection result is that the difference between the first serial number and the second serial number is less than the first difference threshold, and the difference between the third serial number and the fourth serial number is less than the second difference threshold.

[0010] Optionally, determining the target detection result based on the parallel road feature results of the detection points corresponding to the multiple detection lines includes: based on the parallel road feature results of the detection points corresponding to the multiple detection lines, counting the number of detection points continuously having parallel road features on the geometric representation line; detecting whether the number reaches a predetermined threshold; and in the case where the detection result is that the number reaches the predetermined threshold, determining that the target detection result is that there is a parallel road of the target main road.

[0011] According to another aspect of the present disclosure, there is provided an apparatus for detecting a parallel road, including: an acquisition module configured to acquire a geometric representation line corresponding to the passing direction of a target main road and a target trajectory, where the target trajectory is a trajectory passing through the geometric representation line; a first generation module configured to generate a trajectory grid map according to the geometric representation line and the target trajectory, where each grid in the trajectory grid map is marked with the number of target trajectories passing through it; a second generation module configured to select a plurality of detection points from the geometric representation line and generate a plurality of detection lines corresponding to the plurality of detection points; an extraction module configured to map the plurality of detection lines into the trajectory grid map to obtain a target grid map, and extract pixel features of the plurality of detection lines from the target grid map, where the pixel features of the plurality of detection lines are respectively characterized by the number of target trajectories passing through the grids on the corresponding detection lines; and a determination module configured to determine a target detection result based on the pixel features of the plurality of detection lines, where the target detection result is used to characterize whether there is a parallel road of the target main road.

[0012] Optionally, the first generation module includes: a first initial unit configured to generate an initial grid map with a predetermined width using the geometric representation line as the center line; a first acquisition unit configured to respectively acquire the number of target trajectories passing through each grid in the initial grid map; and a first generation unit configured to mark the number of target trajectories passing through each grid in the initial grid map into the corresponding grid to obtain the trajectory grid map.

[0013] Optionally, the second generation module includes: a second initial unit configured to select a plurality of detection points on the geometric representation line based on a predetermined interval; and a second generation unit configured to respectively generate a plurality of detection lines with a predetermined length perpendicular to the geometric representation line and corresponding to the plurality of detection points with the plurality of detection points as the centers.

[0014] Optionally, the determination module includes: a peak determination unit configured to determine a pixel main peak and a pixel secondary peak of each of the multiple detection lines based on pixel features of the multiple detection lines respectively; a feature determination unit configured to determine a parallel road feature result of a detection point corresponding to each of the multiple detection lines based on the pixel main peak and the pixel secondary peak of the multiple detection lines respectively, where the parallel road feature result is used to identify whether the corresponding detection point has a parallel road feature; and a result determination unit configured to determine the target detection result based on the parallel road feature results of the detection points corresponding to the multiple detection lines.

[0015] Optionally, the peak determination unit includes: a main peak determination unit configured to determine pixel features of detection points corresponding to each of the multiple detection lines as the pixel main peak of the corresponding detection line respectively; a clustering unit configured to cluster pixel features on the right side of the pixel main peak corresponding to each of the multiple detection lines respectively to obtain multiple candidate secondary peaks of the corresponding detection line; and a secondary peak determination unit configured to select, from the multiple candidate secondary peaks of the corresponding detection line, a candidate secondary peak with the largest pixel feature as the pixel secondary peak.

[0016] Optionally, the feature determination unit includes: a first numbering unit configured to number the multiple detection points selected on the geometric representation line in ascending order; a second numbering unit configured to cluster pixel features on the right side of the pixel main peak corresponding to each of the multiple detection lines respectively to obtain multiple candidate secondary peaks of the corresponding detection line, and number the multiple candidate secondary peaks in ascending order starting from the detection point; a second obtaining unit configured to, for a target detection point corresponding to a target detection line among the multiple detection lines, obtain a first serial number of the pixel main peak on the target detection line, and a second serial number of the pixel main peak on the previous detection line of the target detection line in the passing direction; obtain a third serial number of the pixel secondary peak on the target detection line, and a fourth serial number of the pixel secondary peak on the previous detection line of the target detection line in the passing direction; a first detection unit configured to detect whether a difference between the first serial number and the second serial number is less than a first difference threshold, and detect whether a difference between the third serial number and the fourth serial number is less than a second difference threshold; and a determination unit configured to determine that the target detection point has a parallel road feature when the detection result shows that the difference between the first serial number and the second serial number is less than the first difference threshold, and the difference between the third serial number and the fourth serial number is less than the second difference threshold.

[0017] Optionally, the result determination unit includes: a statistics unit configured to count the number of detection points with parallel road features continuously on the geometric representation line based on the parallel road feature results of the detection points corresponding to the multiple detection lines; a second detection unit configured to detect whether the number reaches a predetermined threshold; and a result determination subunit configured to determine that the target detection result is that there is a parallel road of the target main road when the detection result is that the number reaches the predetermined threshold.

[0018] According to another aspect of the present disclosure, there is provided an electronic device, including: at least one processor; and a memory communicatively connected to the at least one processor; wherein, the memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute any one of the above methods.

[0019] According to yet another aspect of the present disclosure, there is provided a non-transitory computer-readable storage medium storing computer instructions, wherein the computer instructions are used to cause the computer to execute any one of the above methods.

[0020] According to still another aspect of the present disclosure, there is provided a computer program product, including a computer program which, when executed by a processor, implements any one of the above methods.

[0021] It should be understood that the content described in this part is not intended to identify the key or important features of the embodiments of the present disclosure, nor is it used to limit the scope of the present disclosure. Other features of the present disclosure will become easily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The drawings are used to better understand the solution and do not constitute a limitation to the present disclosure. Among them:

[0023] Figure 1 is a flowchart of the parallel road detection method provided by an embodiment of the present disclosure;

[0024] Figure 2 is a schematic diagram of the parallel road detection method provided according to an alternative embodiment of the present disclosure;

[0025] Figure 3 is a schematic diagram of extracting the traction main road of the parallel auxiliary road provided according to an alternative embodiment of the present disclosure;

[0026] Figure 4 is a schematic diagram after topological reconstruction of the traction main road is completed provided according to an alternative embodiment of the present disclosure;

[0027] Figure 5 is a schematic diagram of trajectory rasterization provided according to an alternative embodiment of the present disclosure;

[0028] Figure 6 It is a schematic diagram of detection points and detection lines provided according to an alternative embodiment of the present disclosure;

[0029] Figure 7 It is a schematic diagram of a cross-sectional waveform after filtering processing provided according to an alternative embodiment of the present disclosure;

[0030] Figure 8 It is schematic data of parallel road features provided according to an alternative embodiment of the present disclosure;

[0031] Figure 9 It is a schematic diagram of parallel road feature points provided according to an alternative embodiment of the present disclosure;

[0032] Figure 10 It is a structural block diagram of a parallel road detection device provided according to an embodiment of the present disclosure;

[0033] Figure 11 It shows a schematic block diagram of an exemplary electronic device 1100 that can be used to implement the embodiments of the present disclosure. Specific Embodiments

[0034] The following describes exemplary embodiments of the present disclosure with reference to the accompanying drawings. Various details of the embodiments of the present disclosure are included to assist understanding, and they should be considered merely exemplary. Therefore, those of ordinary skill in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Similarly, for clarity and conciseness, descriptions of well-known functions and structures are omitted in the following description.

[0035] Term Explanation

[0036] A parallel road refers to a road parallel to a large road in the field of maps, and in urban roads, it is also called the access road of the main road.

[0037] In an embodiment of the present disclosure, a parallel road detection method is provided. Figure 1 It is a flowchart of the parallel road detection method provided by an embodiment of the present disclosure. As Figure 1 shown, the method includes:

[0038] Step S102, obtaining a geometric representation line corresponding to the passing direction of the target main road and a target trajectory, where the target trajectory is a trajectory passing through the geometric representation line;

[0039] Step S104, generating a trajectory grid map according to the geometric representation line and the target trajectory, where each grid in the trajectory grid map is marked with the number of target trajectories passing through;

[0040] Step S106, selecting a plurality of detection points from the geometric representation line and generating a plurality of detection lines corresponding to the plurality of detection points;

[0041] Step S108: Map multiple detection lines into the trajectory grid map to obtain the target grid map, and extract the pixel features of the multiple detection lines from the target grid map, where the pixel features of the multiple detection lines are respectively characterized by the number of target trajectories passing through the grids on the corresponding detection lines.

[0042] Step S110: Determine the target detection result based on the pixel features of the multiple detection lines, where the target detection result is used to characterize whether there is a parallel road to the target main road.

[0043] Through the above steps, the geometric representation line corresponding to the target main road in the traffic direction can be determined. At the same time, the trajectory data related to the target main road can be obtained, and the trajectory grid map and multiple detection lines can be generated by using the obtained geometric representation line and the target trajectory. The trajectory performance of the target trajectory within a certain range of the main road and its vicinity can be statistically obtained in the form of pixel features. Furthermore, it can be directly, efficiently, and accurately determined whether there is a parallel road to the target main road within the predetermined range through the above statistical results. Moreover, this method has a low cost, a high detection result accuracy, and a fast speed, and is suitable for large-scale promotion.

[0044] It should be noted that when determining the geometric representation line according to the target main road, two geometric representation lines can be obtained according to the traffic direction of the target main road, and the parallel roads on both sides of the target main road can be detected according to the two different geometric representation lines. At the same time, when obtaining the geometric representation line, certain preprocessing can be performed according to the actual situation of the road. For example, the short geometric representation lines between intersections can be fitted into a long geometric representation line to avoid the interference of too many small T-shaped intersections; or the geometric representation line can be disconnected at a large intersection to avoid the interference caused by the overly messy trajectories at the intersection; or the geometric representation line can be used to perform topological reconstruction on the main road network, and so on.

[0045] It should be noted that multiple algorithms can be used to achieve the purpose of obtaining the target trajectory. After obtaining the target trajectory, the target trajectory can also be screened to ensure the availability and effectiveness of the trajectory data. For example, inferior trajectories, non-vehicle trajectories, abnormal point trajectories, etc. can be excluded. Other screening or related processing for the target trajectory all belong to the content of the embodiments of the present disclosure and will not be exemplified one by one here.

[0046] As an alternative embodiment, according to the geometric representation line and the target trajectory, a trajectory grid map can be generated in the following manner: taking the geometric representation line as the center line, generating an initial grid map with a predetermined width; respectively obtaining the number of target trajectories passing through each grid in the initial grid map; marking the number of target trajectories passing through each grid in the initial grid map into the corresponding grid to obtain the trajectory grid map. Through the above operations, the passing situation of the target trajectory within a predetermined range near the target main road can be directly counted and obtained in the form of a grid map, so as to facilitate further processing of the target trajectory data. At the same time, the method of counting the target trajectory data is also uncertain. For example, the number of target trajectories passing through each grid can be recorded as a value, or the pixels of each grid can be further assigned with this value to obtain a grayscale map that can intuitively represent the situation of the target trajectory, and so on.

[0047] It should be noted that for the generated trajectory grid map, further processing can be performed to optimize the effect of the trajectory grid map. For example, multi-layer Gaussian filtering processing can be adopted to make the trajectory grid map more uniform and smooth, remove burrs and noise, reduce data interference, and so on. A series of other processing operations on the trajectory grid map all belong to the content of the embodiments of the present disclosure, and will not be exemplified one by one here.

[0048] As an alternative embodiment, multiple detection points are selected from the geometric representation line, and multiple detection lines corresponding to the multiple detection points can be generated in the following manner: multiple detection points are selected on the geometric representation line based on a predetermined interval; respectively taking the multiple detection points as the centers, multiple detection lines with a predetermined length perpendicular to the geometric representation line and corresponding to the multiple detection points are generated. By setting multiple detection points at a predetermined interval and generating multiple detection lines with a predetermined length based on the detection points, the situation of the target trajectory within a certain range near the target main road can be obtained through the multiple detection lines, and then the situation of the parallel road near the target main road can be inferred.

[0049] As an alternative embodiment, based on the pixel features of multiple detection lines, the target detection result can be determined in the following manner: respectively determine the pixel main peaks and pixel secondary peaks of multiple detection lines based on the pixel features of multiple detection lines; based on the pixel main peaks and pixel secondary peaks of multiple detection lines, respectively determine the parallel road feature results of the detection points corresponding to multiple detection lines, where the parallel road feature results are used to identify whether the corresponding detection points have parallel road features; based on the parallel road feature results of the detection points corresponding to multiple detection lines, determine the target detection result. The pixel main peaks and pixel secondary peaks determined from the pixel features of the detection lines can be used to characterize the position of the main road and the possible positions where parallel roads may exist. Based on the pixel main peaks and pixel secondary peaks, the parallel road feature results of the corresponding detection points can be determined, and then the possibility of the existence of a parallel road can be further determined. According to the parallel road feature situation of the detection points, the detection result of the parallel road can be accurately obtained.

[0050] As an alternative embodiment, to respectively determine the pixel main peaks and pixel secondary peaks of multiple detection lines based on the pixel features of multiple detection lines, the following method can be adopted: respectively determine the pixel features of the detection points corresponding to multiple detection lines as the pixel main peaks of the corresponding detection lines; respectively cluster the pixel features on the right side of the corresponding pixel main peaks of multiple detection lines to obtain multiple candidate secondary peaks of the corresponding detection lines; respectively select the candidate secondary peak with the largest pixel feature from the multiple candidate secondary peaks of the corresponding detection line as the pixel secondary peak. The pixel main peak corresponding to the detection line represents the position of the target main road. Since the parallel road is on the right side of the main road when facing the traffic direction, the detection of the parallel road can be realized by further detecting the pixel secondary peaks on the right side of the pixel main peak. By clustering the pixel features on the right side of the pixel main peak, several pixel secondary peaks with obvious wave peaks can be formed, improving the efficiency of determining the pixel secondary peak. Further, the one with the largest peak value, that is, the one with the most times the target trajectory passes through, is determined from several pixel secondary peaks, and it can be considered that there may be a parallel road at this pixel secondary peak.

[0051] It should be noted that in the above process, not only can the pixel secondary peaks be clustered, but also the pixel main peaks can be clustered to reduce unnecessary data interference. For example, when the secondary peak is too close to the main peak, the secondary peak can be merged into the main peak, that is, the secondary peak may still be formed by the trajectory passing through the main road. The distance criterion for determining whether to merge into the main peak is uncertain and can be adjusted according to the actual application situation.

[0052] It should be noted that in the above process, the pixel features can also be subjected to feature enhancement processing. For example, smoothing processing can be performed to enhance the wave peak feature value, that is, the influence caused by the phenomenon of non-concentrated traffic flow due to too wide a road or too many lanes is suppressed. In addition, there are many implementation methods for feature enhancement processing of pixels, all of which belong to the content of the embodiments of the present disclosure and will not be exemplified one by one here.

[0053] It should be noted that when determining the pixel sub-peak with the largest peak value from several candidate sub-peaks, the bubble algorithm can be used.

[0054] As an alternative embodiment, based on the pixel main peaks and pixel sub-peaks of multiple detection lines, the parallel road feature results corresponding to the detection points of multiple detection lines are determined respectively, including: sequentially numbering multiple detection points selected on the geometric representation line; respectively clustering the pixel features on the right side of the pixel main peaks corresponding to multiple detection lines to obtain multiple candidate sub-peaks corresponding to the detection lines, and sequentially numbering the multiple candidate sub-peaks starting from the detection points; for the target detection point corresponding to the target detection line among multiple detection lines, obtaining the first serial number of the pixel main peak on the target detection line, and the second serial number of the pixel main peak on the previous detection line of the target detection line in the traffic direction; obtaining the third serial number of the pixel sub-peak on the target detection line, and the fourth serial number of the pixel sub-peak on the previous detection line of the target detection line in the traffic direction; detecting whether the difference between the first serial number and the second serial number is less than the first difference threshold, and detecting whether the difference between the third serial number and the fourth serial number is less than the second difference threshold; in the case where the detection result is that the difference between the first serial number and the second serial number is less than the first difference threshold, and the difference between the third serial number and the fourth serial number is less than the second difference threshold, it is determined that the target detection point has the parallel road feature. By detecting whether the serial number difference is lower than the predetermined difference threshold, it can be determined whether the pixel main peaks and pixel sub-peaks between detection points can be regarded as continuous. Further, if there is a pixel sub-peak on the detection line of the detection point, and the pixel main peaks and pixel sub-peaks of this detection point and another detection point on the detection line are continuous, this detection point can be regarded as having the parallel road feature, that is, at the pixel sub-peak on the detection line of this detection point, there may be a parallel road.

[0055] As an alternative embodiment, based on the parallel road feature results corresponding to the detection points of multiple detection lines, the target detection result is determined, including: based on the parallel road feature results corresponding to the detection points of multiple detection lines, counting the number of detection points that continuously have the parallel road feature on the geometric representation line; detecting whether the number reaches the predetermined threshold; in the case where the detection result is that the number reaches the predetermined threshold, determining that the target detection result is that there is a parallel road of the target main road. If multiple consecutive detection points have the parallel road feature, it means that the road sections at the pixel sub-peaks on the detection lines of these multiple detection points are continuous, and further, this road section can be determined as a parallel road, that is, the parallel road detection result is obtained. At the same time, according to the numbering of the detection points, it can also be directly determined which detection points detected the parallel road, that is, the position and road section of the parallel road can be directly and efficiently determined.

[0056] It should be noted that when counting the number of detection points with the characteristics of parallel roads in the above-mentioned continuous manner, considering the calculation error of pixel characteristics, if a certain detection point does not have the characteristics of parallel roads, but its two adjacent detection points both have the characteristics of parallel roads, then this detection point is also considered to have the characteristics of parallel roads.

[0057] Based on the above embodiments and optional embodiments, an optional implementation manner is provided.

[0058] In the field of maps, parallel roads refer to the roads parallel to large roads. In urban roads, they are also called the feeder roads of the main roads. In the road planning of many cities, in order to reduce the number of intersections or traffic lights on the main roads and improve the traffic capacity of the main roads, feeder roads parallel to them are often built. Therefore, parallel roads can not only share part of the traffic flow on the main roads, but also play a role in connecting the road network, and are indispensable roads for mapping.

[0059] In the related art, the methods for making parallel roads mainly adopt the following two methods:

[0060] (1) The collection vehicle conducts full-coverage collection of the city. When a new parallel road is found, it collects the main road and the parallel road in the forward direction. Since its positioning equipment belongs to high-precision professional equipment, the trajectories of the main road and the parallel road collected are easy to be distinguished and extracted through technology, and after extraction, they are handed over to a professional team for mapping;

[0061] (2) Users report new parallel roads.

[0062] However, the above two methods have the following disadvantages respectively:

[0063] (1) The cost of collection vehicles is relatively high and the number is limited. It is impossible to complete the national collection in a short time and it is impossible to timely supplement new parallel roads;

[0064] (2) The parallel roads reported by users are very few, and users themselves are not professional teams, so they cannot accurately describe the parallel road scenarios, and the recall rate is too low, which can only play an individual supplementary role and cannot be scaled up.

[0065] The optional implementation manner of the present disclosure proposes a brand-new method for discovering parallel roads without the need for collection vehicles to collect. Using road trajectory data, new parallel roads are discovered through trajectory mining technology, and new parallel roads across the country can be discovered in a short time. Figure 2 It is a schematic diagram of a parallel road detection method provided according to the optional implementation manner of the present disclosure, as Figure 2 shown. The specific scheme is as follows:

[0066] (1) Step 1, extraction and processing of the spatial geometry of the main road.

[0067] Parallel roads are the feeder roads of the main road. Taking the main road as the traction, it can be judged whether there are new parallel roads. Figure 3It is a schematic diagram of extracting the main road that pulls the parallel side road according to an optional implementation manner of the present disclosure. As Figure 3 shown, the screening principle of the main road is a large road with an obvious separator or double yellow line in the middle. When making data mapping, two geometric lines will be drawn according to the driving direction, and each geometric line corresponds to the actual traffic direction of the road. After the geometric lines of the main road are extracted, in order to avoid the interference of too many small T-junctions, it is necessary to perform topological reconstruction on the road network of the main road. For example, the small segments of geometric lines between intersections are fitted into a long geometric line for mining the spatial position traction of parallel roads. At the same time, the geometric lines are interrupted at larger intersections to prevent the interference of messy trajectories at intersections on mining. Figure 4 It is a schematic diagram after the topological reconstruction of the main road that pulls is completed according to an optional implementation manner of the present disclosure. As Figure 4 shown, it is disconnected at large intersections, and part of the geometric lines at the intersections are subtracted to prevent the interference of messy trajectories at the intersections on the scheme effect. The scenarios participating in the topological reconstruction include T-junctions, interruptions such as cameras and speed limit signs, interchanges, and special roads such as bridges and tunnels.

[0068] (2) Step two, trajectory extraction and rasterization.

[0069] Figure 5 It is a schematic diagram of trajectory rasterization according to an optional implementation manner of the present disclosure. As Figure 5 shown, the gray lines are the geometric lines of the main road, and the black lines are the trajectories. The main road is split into two geometric lines according to two driving directions, and each geometric line corresponds to a unique driving direction. The trajectory data bound to the geometric line is extracted, and it is required that the trajectory is a through-trajectory of the geometric line and in the same direction as the geometric line to ensure that the distribution of trajectory points on the geometric line is uniform, so as to improve the effectiveness of the technology. The time interval for trajectory selection is N days, and N depends on the user throughput of the geometric line. N is at least greater than 1 to ensure the randomness of user trajectories. The cumulative bound trajectory volume PV in N days is greater than pv min and less than pv max , and pv max does not exceed 1000. For the extracted trajectories, first perform high-quality trajectory screening to exclude inferior trajectories, non-vehicle driving trajectories, and abnormal point trajectories to ensure the usability of the trajectories. After completion, calculate the road surface geometry with a width of 30 meters for the geometric line and cut it into 1-meter * 1-meter grids. The road surface grids near the intersections need to be excluded. Connect all the trajectory points on the geometric line into a line, and calculate the spatial relationship between the line and the grid. When a certain grid is passed by a certain trajectory, the count is incremented by 1, and the same calculation is performed for all trajectories. It can be obtained how many times each grid is passed by the trajectory. Let this value be v, and v can be used as the value of the raster image pixel, and then the trajectory raster map of the geometric line can be obtained. Perform multi-layer Gaussian filtering on this trajectory raster map to make it uniform and smooth, and remove burrs and noise.

[0070] (3) Step three, extract the feature detection line.

[0071] The parallel road feature information is hidden in the geometric line grid map, and the features show a discrete characteristic. Figure 6 It is a schematic diagram of detection points and detection lines provided according to an alternative embodiment of the present disclosure. As Figure 6 shown, in order to extract these discrete feature points and retain a certain continuity, in the alternative embodiment of the present disclosure, dotting processing is performed on the one-way geometric line of the main road, with a dot being made every 20 meters, and each dot serves as a feature detection point. A perpendicular line is calculated with the detection point as the foot of the perpendicular, and the length of the perpendicular line does not exceed 50 meters, which is used as the feature detection line. Since the intersection trajectory is too messy, to ensure the technical effect, the detection lines at the intersection can be deleted. At the same time, the feature detection lines are sorted and numbered based on the direction of the geometric line shape points to ensure that when calculating features, the ascending order of the detection line numbers is consistent with the trajectory direction. And the geometric generation method of the detection line is centered on the geometric line direction and from left to right.

[0072] (4) Step Four, extracting the cross-sectional wave features based on the detection lines.

[0073] After the feature detection lines are extracted in Step Three, in the same way as mapping the grid in Step Two, the feature detection lines are mapped into the trajectory grayscale map, and a sequence of pixel points [p1, p2…p n , p (n+1) , …p (2n-1) is obtained, where p n is the pixel point at the midpoint of the detection line, that is, the intersection point of the detection line and the geometric line. Based on the mapping relationship between this sequence of pixel points and the trajectory grayscale map, a sequence of pixel values [V1, V2, …V n , V (n+1) , …V (2n-1) is obtained. This sequence is called the one-dimensional feature vector of the geometric line based on this point, denoted as vector A. According to the sequence order of the detection points of the geometric line, the one-dimensional feature vectors of each detection point are obtained, and the two-dimensional feature vector of this geometric line is constructed, denoted as B, that is, B = [A1, A2, A3…A4]. Figure 7 It is a schematic diagram of the cross-sectional waveform after filtering processing provided according to an alternative embodiment of the present disclosure.

[0074] (5) Step Five, fitting the parallel road features based on the multi-point cross-sectional wave features.

[0075] The two-dimensional feature vector of the one-way geometric line of the main road is obtained in Step Four. To extract more effective parallel road features, feature enhancement processing needs to be performed on this two-dimensional vector. First, for the one-dimensional feature vector [V1, V2, …V n , V (n+1) , …V (2n-1)Perform feature enhancement processing. The processing steps are to smooth it using one-dimensional Gaussian filtering and enhance the peak feature values. This step can significantly suppress the parallel characteristics of vehicle flow caused by overly wide roads or excessive lanes. Use the peak bubbling algorithm to quickly find the global peak of this one-dimensional vector, denoted as point m, and its value is V m . According to the main road traffic characteristics, this point m is the main road peak. Since the parallel road exists on the right side of the geometric line driving direction, the value of point x with a point sequence less than m can be normalized to 0, that is, V x = 0 (x < m). Then process the secondary peak characteristics on the right side of the geometric line. Use the DBSCAN clustering algorithm to further aggregate several small peaks on the right side of the peak to form several obvious secondary peaks. If the distance between the secondary peak and the main peak is less than 4 meters, it is merged into the main peak. Let the peak value be P, and obtain the one-dimensional parallel road feature vector [0, 0…0, P m , 0…0, P m1 , 0…P m2 , 0…], where p m is the main peak, p m1 is the first secondary peak, p m2 is the second secondary peak. Select Max(p m1 , p m2 ) as the parallel road peak point. Perform the above processing on the one-dimensional feature vectors of all detection points on the geometric line to obtain the two-dimensional parallel road feature vector P A .

[0076] (6) Step six, extract parallel road information based on the parallel road feature vector.

[0077] According to the parallel road feature vector extracted in step five, further calculate the parallel road feature points. For the one-dimensional feature vector corresponding to each detection point, if it meets the following conditions, it is considered that this detection point has parallel road characteristics: 1) The difference between the main peak point serial number of this detection point and the main peak point serial number value of the previous detection point is within 2; 2) This detection point has a secondary peak, and the difference between the secondary peak serial number value and the secondary peak serial number value of the previous detection point is within 2. The detection points that meet the above two conditions are considered to have parallel road characteristics. Calculate the parallel road characteristics for all detection points on the geometric line. Let the above method be F, and obtain the parallel road feature vector P L , that is, P L = F(P A ). For example, if there are 8 detection points on the geometric line, and the points numbered 2, 3, 4, and 6 have parallel road characteristics, then the feature form is [v0, v1, v1, v1, v0, v1, v0, v0], where v0 represents no parallel road characteristics and v1 represents having parallel road characteristics. For P LFurther interpolation processing is performed. If there are parallel road features before and after v0, then v0 is set to v1. Connect all the detection points with the value of v1 on the geometric line to form a survey line geometry, and if the number of detection points constituting the survey line exceeds 3 points, that is, it has the continuity feature, then it can be regarded as the parallel road traction survey line of the geometric line, that is, there is a parallel road in this section.

[0078] Figure 8 is the schematic data of parallel road features provided according to an alternative embodiment of the present disclosure, such as Figure 8 shown, "1" in the second column indicates no parallel road feature, and "2" indicates there is a parallel road feature. The fourth column is the detection point serial number. The third column in the second row [(2,775),(19,119)] indicates that there is a peak at the position with subscript 2 in the feature vector of the 40th detection point, with a value of 775, and there is a secondary peak at the position with subscript 19, with a value of 119. It can be seen from the schematic data that detection points 40 to 46 can form a parallel road.

[0079] Figure 9 is the schematic diagram of parallel road feature points provided according to an alternative embodiment of the present disclosure, such as Figure 9 shown, the points in the figure are parallel road feature points, showing obvious continuity features, which have a significant effect on discovering parallel roads.

[0080] An alternative embodiment of the present disclosure proposes a new method for discovering parallel roads. Compared with the methods in related technologies, this technology has the advantages of high timeliness and high recall. According to the experimental data, the alternative embodiment of the present disclosure can discover newly opened medium and large parallel roads nationwide within two days, and small parallel auxiliary roads nationwide within one week. The discovery timeliness of parallel roads nationwide has been shortened from monthly or even yearly to daily level, greatly reducing the acquisition cost of high-precision equipment, and can be applied to the real road network mining system, becoming the main discovery means of parallel roads.

[0081] In an embodiment of the present disclosure, a device for parallel road detection is also provided. Figure 10 is the structural block diagram of the parallel road detection device provided according to an embodiment of the present disclosure, such as Figure 10 shown, the device includes: an acquisition module 1001, a first generation module 1002, a second generation module 1003, an extraction module 1004, and a determination module 1005. The device will be described below:

[0082] An acquisition module 1001, configured to acquire a geometric representation line corresponding to a target main road passing direction and a target trajectory, where the target trajectory is a trajectory passing through the geometric representation line; a first generation module 1002, connected to the acquisition module 1001, configured to generate a trajectory grid map according to the geometric representation line and the target trajectory, where each grid in the trajectory grid map is marked with the number of target trajectories passing through; a second generation module 1003, connected to the first generation module 1002, configured to select a plurality of detection points from the geometric representation line and generate a plurality of detection lines corresponding to the plurality of detection points; an extraction module 1004, connected to the second generation module 1003, configured to map the plurality of detection lines into the trajectory grid map to obtain a target grid map, and extract pixel features of the plurality of detection lines from the target grid map, where the pixel features of the plurality of detection lines are respectively characterized by the number of target trajectories passing through the grids on the corresponding detection lines; a determination module 1005, connected to the extraction module 1004, configured to determine a target detection result based on the pixel features of the plurality of detection lines, where the target detection result is used to characterize whether there is a parallel road of the target main road.

[0083] As an optional embodiment, the first generation module 1002 includes: a first initial unit, configured to generate an initial grid map with a predetermined width using the geometric representation line as the center line; a first acquisition unit, configured to respectively acquire the number of target trajectories passing through each grid in the initial grid map; a first generation unit, configured to mark the number of target trajectories passing through each grid in the initial grid map to the corresponding grid to obtain a trajectory grid map.

[0084] As an optional embodiment, the second generation module 1003 includes: a second initial unit, configured to select a plurality of detection points on the geometric representation line based on a predetermined interval; a second generation unit, configured to respectively generate a plurality of detection lines perpendicular to the geometric representation line and corresponding to the plurality of detection points with a predetermined length centered on the plurality of detection points.

[0085] As an optional embodiment, the determination module 1005 includes: a peak determination unit, configured to respectively determine the pixel main peak and pixel sub-peak of the plurality of detection lines based on the pixel features of the plurality of detection lines; a feature determination unit, configured to respectively determine the parallel road feature results of the detection points corresponding to the plurality of detection lines based on the pixel main peak and pixel sub-peak of the plurality of detection lines, where the parallel road feature results are used to identify whether the corresponding detection points have parallel road features; a result determination unit, configured to determine the target detection result based on the parallel road feature results of the detection points corresponding to the plurality of detection lines.

[0086] As an alternative embodiment, the peak determination unit includes: a main peak determination unit for respectively determining the pixel features of the corresponding detection points on multiple detection lines as the pixel main peaks of the corresponding detection lines; a clustering unit for respectively clustering the pixel features on the right side of the corresponding pixel main peaks on multiple detection lines to obtain multiple candidate secondary peaks of the corresponding detection lines; and a secondary peak determination unit for respectively selecting the candidate secondary peak with the largest pixel feature from the multiple candidate secondary peaks of the corresponding detection line as the pixel secondary peak.

[0087] As an alternative embodiment, the feature determination unit includes: a first numbering unit for sequentially numbering the multiple detection points selected on the geometric representation line in ascending order; a second numbering unit for respectively clustering the pixel features on the right side of the corresponding pixel main peaks on multiple detection lines to obtain multiple candidate secondary peaks of the corresponding detection lines, and sequentially numbering the multiple candidate secondary peaks in ascending order starting from the detection points; a second acquisition unit for, for the target detection point corresponding to the target detection line among multiple detection lines, acquiring the first serial number of the pixel main peak on the target detection line, and the second serial number of the pixel main peak on the previous detection line of the target detection line in the passing direction; acquiring the third serial number of the pixel secondary peak on the target detection line, and the fourth serial number of the pixel secondary peak on the previous detection line of the target detection line in the passing direction; a first detection unit for detecting whether the difference between the first serial number and the second serial number is less than a first difference threshold, and detecting whether the difference between the third serial number and the fourth serial number is less than a second difference threshold; and a determination unit for determining that the target detection point has a parallel road feature when the detection result is that the difference between the first serial number and the second serial number is less than the first difference threshold, and the difference between the third serial number and the fourth serial number is less than the second difference threshold.

[0088] As an alternative embodiment, the result determination unit includes: a statistics unit for counting the number of detection points continuously having parallel road features on the geometric representation line based on the parallel road feature results of the corresponding detection points on multiple detection lines; a second detection unit for detecting whether the number reaches a predetermined threshold; and a result determination subunit for determining that the target detection result is the existence of a parallel road of the target main road when the detection result is that the number reaches the predetermined threshold.

[0089] In the technical solution of the present disclosure, the acquisition, storage, and application of the user's personal information involved all comply with the provisions of relevant laws and regulations and do not violate public order and good customs.

[0090] According to the embodiments of the present disclosure, the present disclosure also provides an electronic device, a readable storage medium, and a computer program product.

[0091] Figure 11FIG. shows a schematic block diagram of an exemplary electronic device 1100 that can be used to implement embodiments of the present disclosure. The electronic device is intended to represent various forms of digital computers, such as, for example, laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as, for example, personal digital processors, cellular telephones, smart phones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely exemplary and are not intended to limit the implementations of the present disclosure described and / or claimed herein.

[0092] As Figure 11 shown, the device 1100 includes a computing unit 1101 that can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 1102 or a computer program loaded from a storage unit 1108 into a random access memory (RAM) 1103. In the RAM 1103, various programs and data required for the operation of the device 1100 can also be stored. The computing unit 1101, the ROM 1102, and the RAM 1103 are connected to each other via a bus 1104. An input / output (I / O) interface 1105 is also connected to the bus 1104.

[0093] A plurality of components in the device 1100 are connected to the I / O interface 1105, including: an input unit 1106, such as, for example, a keyboard, a mouse, etc.; an output unit 1107, such as, for example, various types of displays, speakers, etc.; a storage unit 1108, such as, for example, a magnetic disk, an optical disk, etc.; and a communication unit 1109, such as, for example, a network card, a modem, a wireless communication transceiver, etc. The communication unit 1109 allows the device 1100 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0094] The computing unit 1101 can be various general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 1101 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various dedicated artificial intelligence (AI) computing chips, various computing units running machine learning model algorithms, a digital signal processor (DSP), and any suitable processor, controller, microcontroller, etc. The computing unit 1101 executes the various methods and processes described above, such as the parallel road detection method. For example, in some embodiments, the parallel road detection method can be implemented as a computer software program tangibly embodied in a machine-readable medium, such as the storage unit 1108. In some embodiments, part or all of the computer program can be loaded and / or installed onto the device 1100 via the ROM 1102 and / or the communication unit 1109. When the computer program is loaded into the RAM 1103 and executed by the computing unit 1101, one or more steps of the parallel road detection method described above can be executed. Alternatively, in other embodiments, the computing unit 1101 can be configured to execute the parallel road detection method by any other suitable means (e.g., by means of firmware).

[0095] The various embodiments of the systems and techniques described above in this document can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGA), application-specific integrated circuits (ASIC), application-specific standard products (ASSP), systems-on-a-chip (SOC), complex programmable logic devices (CPLD), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include: being implemented in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which can be a special or general programmable processor, and can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit the data and instructions to the storage system, the at least one input device, and the at least one output device.

[0096] The program code for implementing the methods of the present disclosure can be written in any combination of one or more programming languages. These program codes can be provided to a processor or controller of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the program code is executed by the processor or controller, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The program code can be executed entirely on the machine, partially on the machine, as an independent software package partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0097] In the context of this disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in connection with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can include, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of a machine-readable storage medium would include an electrical connection based on one or more wires, a portable computer diskette, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or Flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0098] In order to provide interaction with a user, the systems and techniques described herein can be implemented on a computer having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and a pointing device (e.g., a mouse or a trackball) by which the user can provide input to the computer. Other kinds of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0099] The systems and techniques described herein can be implemented in a computing system including backend components (e.g., as a data server), or a computing system including middleware components (e.g., an application server), or a computing system including frontend components (e.g., a user computer having a graphical user interface or a web browser through which the user can interact with an implementation of the systems and techniques described herein), or a computing system including any combination of such backend components, middleware components, or frontend components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), and the Internet.

[0100] A computer system can include a client and a server. The client and the server are generally remote from each other and typically interact through a communication network. The client-server relationship is generated by computer programs running on the respective computers and having a client-server relationship with each other. The server can be a cloud server, can also be a server of a distributed system, or a server incorporating a blockchain.

[0101] It should be understood that the various forms of processes shown above can be used, with steps reordered, added or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solutions disclosed in this disclosure can be achieved, and no limitation is imposed herein.

[0102] The above specific embodiments do not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub - combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure shall be included within the scope of protection of this disclosure.

Claims

1. A method for detecting parallel roads, comprising: Obtain the geometric representation line corresponding to the target main road traffic direction and the target trajectory, where the target trajectory is a trajectory passing through the geometric representation line, and the geometric representation line is a line drawn according to the driving direction and representing the actual traffic direction of the road; Generate a trajectory grid map based on the geometric representation line and the target trajectory, where each grid in the trajectory grid map is marked with the number of target trajectories passing through it; Select multiple detection points from the geometric representation line and generate multiple detection lines corresponding to the multiple detection points; Map the multiple detection lines into the trajectory grid map to obtain a target grid map, and extract the pixel features of the multiple detection lines from the target grid map, where the pixel features of the multiple detection lines are respectively characterized by the number of target trajectories passing through the grids on the corresponding detection lines; Determine a target detection result based on the pixel features of the multiple detection lines, where the target detection result is used to characterize whether there is a parallel road to the target main road; Among them, the step of selecting multiple detection points from the geometric representation line and generating multiple detection lines corresponding to the multiple detection points includes: selecting multiple detection points on the geometric representation line based on a predetermined interval; respectively taking the multiple detection points as the centers, and generating multiple detection lines perpendicular to the geometric representation line and corresponding to the multiple detection points with a predetermined length.

2. The method according to claim 1, wherein, The step of generating a trajectory grid map based on the geometric representation line and the target trajectory includes: Generate an initial grid map with a predetermined width using the geometric representation line as the center line; Respectively obtain the number of target trajectories passing through each grid in the initial grid map; Mark the number of target trajectories passing through each grid in the initial grid map into the corresponding grid to obtain the trajectory grid map.

3. The method according to claim 1, wherein, The step of determining a target detection result based on the pixel features of the multiple detection lines includes: Respectively determine the pixel main peak and pixel secondary peak of the pixel features of the multiple detection lines based on the pixel features of the multiple detection lines; Based on the pixel main peak and pixel secondary peak of the multiple detection lines, respectively determine the parallel road feature results of the detection points corresponding to the multiple detection lines, where the parallel road feature results are used to identify whether the corresponding detection points have parallel road features; Determine the target detection result based on the parallel road feature results of the detection points corresponding to the multiple detection lines.

4. The method according to claim 3, wherein, The step of respectively determining the pixel main peak and pixel secondary peak of the pixel features of the multiple detection lines based on the pixel features of the multiple detection lines includes: Respectively determine that the pixel feature of the detection point corresponding to each of the multiple detection lines is the pixel main peak of the corresponding detection line; Respectively perform clustering on the pixel features on the right side of the corresponding pixel main peak on the multiple detection lines to obtain multiple candidate secondary peaks of the corresponding detection lines; Respectively select the candidate secondary peak with the largest pixel feature from the multiple candidate secondary peaks of the corresponding detection line as the pixel secondary peak.

5. The method according to claim 3, wherein, The step of respectively determining the parallel road feature results of the detection points corresponding to the multiple detection lines based on the pixel main peak and pixel secondary peak of the multiple detection lines includes: Number the multiple detection points selected on the geometric representation line in ascending order; Cluster the pixel features on the right side of the pixel main peak corresponding to each of the multiple detection lines to obtain multiple candidate secondary peaks corresponding to the detection lines, and sequentially number the multiple candidate secondary peaks starting from the detection point; For a target detection point corresponding to a target detection line among the multiple detection lines, obtain a first sequence number of the pixel main peak on the target detection line, and a second sequence number of the pixel main peak on the previous detection line of the target detection line in the traffic direction; obtain a third sequence number of the pixel secondary peak on the target detection line, and a fourth sequence number of the pixel secondary peak on the previous detection line of the target detection line in the traffic direction; Detect whether the difference between the first sequence number and the second sequence number is less than a first difference threshold, and detect whether the difference between the third sequence number and the fourth sequence number is less than a second difference threshold; In the case where the detection result is that the difference between the first sequence number and the second sequence number is less than the first difference threshold, and the difference between the third sequence number and the fourth sequence number is less than the second difference threshold, determine that the target detection point has a parallel road feature.

6. The method according to claim 3, wherein, The determining the target detection result based on the parallel road feature results of the detection points corresponding to the multiple detection lines includes: Based on the parallel road feature results of the detection points corresponding to the multiple detection lines, count the number of detection points continuously having parallel road features on the geometric representation line; Detect whether the number reaches a predetermined threshold; In the case where the detection result is that the number reaches the predetermined threshold, determine that the target detection result is that there is a parallel road of the target main road.

7. A device for detecting parallel roads, comprising: An acquisition module, configured to acquire a geometric representation line corresponding to the traffic direction of the target main road and a target trajectory, where the target trajectory is a trajectory passing through the geometric representation line, and the geometric representation line is a line drawn according to the driving direction and representing the actual traffic direction of the road; A first generation module, configured to generate a trajectory grid map according to the geometric representation line and the target trajectory, where each grid in the trajectory grid map is marked with the number of target trajectories passing through; A second generation module, configured to select a plurality of detection points from the geometric representation line and generate a plurality of detection lines corresponding to the plurality of detection points; An extraction module, configured to map the plurality of detection lines into the trajectory grid map to obtain a target grid map, and extract pixel features of the plurality of detection lines from the target grid map, where the pixel features of the plurality of detection lines are respectively characterized by the number of target trajectories passing through the grids on the corresponding detection lines; A determination module, configured to determine a target detection result based on the pixel features of the plurality of detection lines, where the target detection result is used to characterize whether there is a parallel road of the target main road; Wherein, the second generation module includes: a second initial unit, configured to select a plurality of detection points on the geometric representation line based on a predetermined interval; a second generation unit, configured to respectively generate a plurality of detection lines perpendicular to the geometric representation line and having a predetermined length corresponding to the plurality of detection points with the plurality of detection points as the centers.

8. The device according to claim 7, wherein, The first generation module includes: A first initial unit for generating an initial grid map with a predetermined width using the geometric representation line as the center line; A first acquisition unit for respectively acquiring the number of target trajectories passing through each grid in the initial grid map; A first generation unit for marking the number of target trajectories passing through each grid in the initial grid map into the corresponding grid to obtain the trajectory grid map.

9. The device according to claim 7, wherein, The determination module includes: A peak determination unit for respectively determining the pixel main peak and pixel secondary peak of the multiple detection lines based on the pixel features of the multiple detection lines; A feature determination unit for respectively determining the parallel road feature results of the detection points corresponding to the multiple detection lines based on the pixel main peak and pixel secondary peak of the multiple detection lines, where the parallel road feature results are used to identify whether the corresponding detection points have parallel road features; A result determination unit for determining the target detection result based on the parallel road feature results of the detection points corresponding to the multiple detection lines.

10. The device according to claim 9, wherein, The peak determination unit includes: A main peak determination unit for respectively determining that the pixel feature corresponding to the detection point on the multiple detection lines is the pixel main peak of the corresponding detection line; A clustering unit for respectively clustering the pixel features on the right side of the pixel main peak on the multiple detection lines to obtain multiple candidate secondary peaks of the corresponding detection lines; A secondary peak determination unit for respectively selecting the candidate secondary peak with the largest pixel feature from the multiple candidate secondary peaks of the corresponding detection line as the pixel secondary peak.

11. The device according to claim 9, wherein, The feature determination unit includes: A first numbering unit for numbering the multiple detection points selected on the geometric representation line in ascending order; A second numbering unit for respectively clustering the pixel features on the right side of the pixel main peak on the multiple detection lines to obtain multiple candidate secondary peaks of the corresponding detection lines, and numbering the multiple candidate secondary peaks in ascending order starting from the detection point; A second acquisition unit for, for the target detection point corresponding to the target detection line among the multiple detection lines, acquiring the first serial number of the pixel main peak on the target detection line, and the second serial number of the pixel main peak on the previous detection line of the target detection line in the traffic direction; acquiring the third serial number of the pixel secondary peak on the target detection line, and the fourth serial number of the pixel secondary peak on the previous detection line of the target detection line in the traffic direction; A first detection unit for detecting whether the difference between the first serial number and the second serial number is less than a first difference threshold, and detecting whether the difference between the third serial number and the fourth serial number is less than a second difference threshold; A determination unit for determining that the target detection point has parallel road features when the detection result is that the difference between the first serial number and the second serial number is less than the first difference threshold, and the difference between the third serial number and the fourth serial number is less than the second difference threshold.

12. The device according to claim 9, wherein, The result determination unit includes: A statistics unit for counting the number of detection points continuously having parallel road features on the geometric representation line based on the parallel road feature results of the detection points corresponding to the multiple detection lines; A second detection unit for detecting whether the number reaches a predetermined threshold; A result determination subunit, configured to determine, when the detection result indicates that the number reaches the predetermined threshold, that the target detection result is that there is a parallel road to the target main road.

13. An electronic device, comprising: At least one processor; And A memory communicatively connected to the at least one processor; wherein, The memory stores instructions executable by the at least one processor, and when the instructions are executed by the at least one processor, the at least one processor is enabled to execute the method according to any one of claims 1 to 6.

14. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to execute the method according to any one of claims 1 to 6.

15. A computer program product, comprising a computer program which, when executed by a processor, implements the method according to any one of claims 1 to 6.

Images