High-precision map road network matching method, device, electronic device and medium

Abstract

This disclosure provides a road network matching method, relating to the field of autonomous driving technology, and particularly to the field of high-precision maps and road network matching technology. The implementation scheme includes: acquiring target road network data and reference road network data, wherein the target road network data includes multiple target feature points and the reference road network data includes multiple reference feature points; for each target feature point, determining multiple corresponding points to be matched; dividing the multiple points to be matched into multiple levels; obtaining an initial frequency distribution, and performing iterative operations on the multiple levels based on the initial frequency distribution; obtaining the probability matrix output by the iterative operations; performing relaxation iteration on the probability matrix based on the support parameter; and determining the matching relationship between the reference road network data and the target road network data based on the probability matrix obtained from the relaxation iteration.

Description

Technical Field

[0001] This disclosure relates to the field of autonomous driving technology, and more particularly to the field of high-precision map and road network matching technology, specifically to a road network matching method, apparatus, electronic device, computer-readable storage medium, and computer program product. Background Technology

[0002] High-precision maps, also known as high-resolution maps, are used by autonomous vehicles. They possess accurate vehicle location information and rich road element data, helping cars anticipate complex road conditions such as slope, curvature, and heading, thus better avoiding potential risks.

[0003] Due to differences in map scale and map generalization, the same geographic entity can appear differently on maps from different sources. For example, a building might be represented as point data on a small-scale map, but as polygon data on a large-scale map. Establishing matching relationships for the same target within the same area is the first step in multi-source spatial data integration. Based on the matching results for the same geographic entity, data updates can be achieved quickly. Road networks, as the skeleton of a city, play a crucial role in spatial data integration. The complex topological relationships and diverse structures of roads in local areas, along with the high similarity between some different roads, increase the uncertainty of automatic road network matching. Furthermore, road network matching has wide applications. For instance, it can be used to synchronize data information by matching high-precision maps with 2D maps, and the matching results can also be used for human-machine collaborative driving; matching high-precision map data from different versions can detect data changes between versions.

[0004] The methods described in this section are not necessarily methods that had been previously conceived or adopted. Unless otherwise specified, no method described in this section should be assumed to be prior art simply because it is included in this section. Similarly, unless otherwise specified, the issues mentioned in this section should not be considered to be accepted in any prior art. Summary of the Invention

[0005] This disclosure provides a road network matching method, apparatus, electronic device, computer-readable storage medium, and computer program product.

[0006] According to one aspect of this disclosure, a road network matching method is provided, comprising: acquiring target road network data and reference road network data, wherein the target road network data includes a plurality of target feature points, and the reference road network data includes a plurality of reference feature points; for each target feature point among the plurality of target feature points, determining a plurality of matchable points among the plurality of reference feature points corresponding to the target feature point, and dividing the plurality of matchable points into a plurality of levels; acquiring an initial frequency distribution, the initial frequency distribution being used to map a plurality of similarity matching parameters into probability values; performing an iterative operation on the plurality of levels based on the initial frequency distribution, to obtain an updated frequency distribution based on the probability matrix output by each iteration for the next iteration, wherein the object of each iteration operation is the matchable points in the already traversed levels and the levels to be traversed; acquiring a probability matrix output based on the iterative operation; acquiring a support parameter corresponding to the probability matrix, and performing relaxation iteration on the probability matrix based on the support parameter; and determining the matching relationship between the reference road network data and the target road network data based on the probability matrix obtained by the relaxation iteration.

[0007] According to another aspect of this disclosure, a road network matching device is provided, comprising: a first acquisition module configured to acquire target road network data and reference road network data, wherein the target road network data includes a plurality of target feature points, and the reference road network data includes a plurality of reference feature points; a first determination module configured to, for each of the plurality of target feature points, determine a plurality of matchable points corresponding to the target feature point among the plurality of reference feature points, and divide the plurality of matchable points into a plurality of levels; a second acquisition module configured to acquire an initial frequency distribution, the initial frequency distribution being used to map a plurality of similarity matching parameters to probability values; and an iteration module configured to... Based on the initial frequency distribution, iterative operations are performed on the multiple levels to obtain an updated frequency distribution for the next iteration based on the probability matrix output by each iteration. Each iteration operates on the points to be matched in the already traversed levels and the levels to be traversed. A third acquisition module is configured to acquire the probability matrix output by the iterative operations. A fourth acquisition module is configured to acquire the support parameter corresponding to the probability matrix and perform relaxation iteration on the probability matrix based on the support parameter. A second determination module is configured to determine the matching relationship between the reference road network data and the target road network data based on the probability matrix obtained by the relaxation iteration.

[0008] According to another aspect of this disclosure, an electronic device is provided, 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, the instructions being executed by the at least one processor to enable the at least one processor to perform a road network matching method.

[0009] According to another aspect of this disclosure, a non-transitory computer-readable storage medium is provided storing computer instructions, wherein the computer instructions are used to cause the computer to perform a road network matching method.

[0010] According to another aspect of this disclosure, a computer program product is provided, comprising a computer program that, when executed by a processor, implements a road network matching method.

[0011] According to one or more embodiments of this disclosure, a method is provided to establish a probability matrix and perform road network matching based on two sets of input road network data. The probability matrix can be iterated and adjusted step by step according to the input data to adapt to the statistical characteristics of the data and obtain the best matching result. This method can effectively improve the accuracy of road network matching.

[0012] It should be understood that the description in this section is not intended to identify key or essential features of the embodiments of this disclosure, nor is it intended to limit the scope of this disclosure. Other features of this disclosure will become readily apparent from the following description. Attached Figure Description

[0013] The accompanying drawings exemplify embodiments and form part of the specification, serving together with the textual description to explain exemplary implementations of the embodiments. The illustrated embodiments are for illustrative purposes only and do not limit the scope of the claims. Throughout the drawings, the same reference numerals refer to similar but not necessarily identical elements.

[0014] Figure 1 A flowchart of a road network matching method according to an embodiment of the present disclosure is shown;

[0015] Figure 2 A schematic diagram illustrating the hierarchical division of buffer regions according to an embodiment of the present disclosure is shown;

[0016] Figure 3 A schematic diagram of a probability matrix according to an embodiment of the present disclosure is shown;

[0017] Figure 4 A structural block diagram of a road network matching device according to an embodiment of the present disclosure is shown; and

[0018] Figure 5A structural block diagram of an exemplary electronic device that can be used to implement embodiments of the present disclosure is shown. Detailed Implementation

[0019] The exemplary embodiments of this disclosure are described below with reference to the accompanying drawings, including various details of the embodiments to aid understanding, and should be considered merely exemplary. Therefore, those skilled in the art will recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope of this disclosure. Similarly, for clarity and brevity, descriptions of well-known functions and structures are omitted in the following description.

[0020] In this disclosure, unless otherwise stated, the use of terms such as "first," "second," etc., to describe various elements is not intended to limit the positional, temporal, or importance relationships of these elements; such terms are merely used to distinguish one element from another. In some examples, the first element and the second element may refer to the same instance of that element, while in other cases, based on the context, they may refer to different instances.

[0021] The terminology used in the description of the various examples described in this disclosure is for the purpose of describing particular examples only and is not intended to be limiting. Unless the context explicitly indicates otherwise, an element may be one or more unless the number of elements is specifically limited. Furthermore, the term "and / or" as used in this disclosure covers any one of the listed items and all possible combinations thereof.

[0022] In related technologies, buffer surface data can be generated based on the target road network according to a fixed threshold. This surface data is then used to calculate spatial relationships with reference data. If a large proportion of the reference lane falls within the buffer, the match is considered successful; otherwise, the match fails. Alternatively, road network matching can be based on topological relationships, using the number of connections between lanes to determine the matching degree between two lanes. In some schemes, a circular region can be generated using lane shape points. Spatial judgment is then performed on the lane shape points of the target and reference road network data. If a certain proportion of the lanes fall within the circular buffer, the two lanes are considered a match; otherwise, the match fails. However, all of the above schemes rely on empirical thresholds and weights to determine matching. These empirically derived parameters may perform well for specific data but lack universality and have a low degree of automation.

[0023] To address the aforementioned issues, this disclosure provides a method for matching road network data by establishing a probability matrix based on two sets of input road network data. This probability matrix can be iterated and adjusted progressively according to the input data to adapt to the statistical characteristics of the data and obtain the optimal matching result. This method can effectively improve the accuracy of road network matching.

[0024] The embodiments of this disclosure will now be described in detail with reference to the accompanying drawings.

[0025] Figure 1 A flowchart of a road network matching method according to an embodiment of the present disclosure is shown. Figure 1 As shown, the road network matching method 100 includes: Step S101, acquiring target road network data and reference road network data, wherein the target road network data includes multiple target feature points, and the reference road network data includes multiple reference feature points; Step S102, for each target feature point among the multiple target feature points, determining multiple matchable points among the multiple reference feature points corresponding to the target feature point, and dividing the multiple matchable points into multiple levels; Step S103, acquiring an initial frequency distribution, wherein the initial frequency distribution is used to map multiple similarity matching parameters into probability values; Step S104, based on the initial frequency... The distribution is iterated over multiple levels to obtain an updated frequency distribution based on the probability matrix output by each iteration for the next iteration. Each iteration operates on the points to be matched in the already traversed levels and the levels to be traversed. Step S105: Obtain the probability matrix output by the iteration operation. Step S106: Obtain the support parameter corresponding to the probability matrix and perform relaxation iteration on the probability matrix based on the support parameter. Step S107: Determine the matching relationship between the reference road network data and the target road network data based on the probability matrix obtained from the relaxation iteration.

[0026] First, matching points are obtained based on target and reference road network data, and these points are categorized into levels. The initial frequency distribution is used to evaluate the matching probability of the currently traversed points and the next level of matching points. This process is iterated until all levels have been traversed, resulting in a preliminary matching probability matrix. Then, the probability matrix is ​​iteratively updated based on contextual information to obtain a globally optimized probability matrix, which is used to determine the matching relationship between the reference and target road network data. Thus, a probability matrix is ​​established based on the two sets of input road network data, and road network data matching is performed based on this probability matrix. This probability matrix can be iteratively adjusted according to the input data to adapt to the statistical characteristics of the data, obtaining the best matching results and effectively improving the accuracy of road network matching.

[0027] According to some embodiments, step S102 includes: for each of the plurality of target feature points, creating a circular buffer zone in the target road network data with the target feature point as the center and a preset length as the radius; performing overlay analysis in the reference road network data based on the circular buffer zone to determine the buffer area corresponding to the target feature point in the reference road network data, wherein the buffer area is an estimated area where a reference feature point matching the target feature point is located; and determining the plurality of points to be matched corresponding to the target feature point in the buffer area. Thus, by using the buffer area analysis method to obtain points to be matched based on the target road network data and the reference road network data, the size of the buffer area can be determined based on an empirical threshold to initially screen reference feature points, while ensuring that the points to be matched fall within the buffer area. Further analysis of the buffer area can then accurately locate the reference feature point matching the target feature point.

[0028] According to some embodiments, step S102 further includes: dividing the buffer area into multiple levels based on the distance from the center of the buffer area; and determining the level corresponding to each of the multiple points to be matched. The method further includes: for each of the multiple levels, calculating the mean of the topological similarity matching parameters corresponding to the points to be matched in that level as a reliability parameter corresponding to that level; and sorting the multiple levels based on the reliability parameter. It is understood that distance is one of the parameters for measuring whether two points match. Intuitively, the closer two points are, the greater the probability of matching. However, the inventors found that in the entire road network data, the probability of matching between two closer points is not necessarily greater. Therefore, the method uses the mean of the topological similarity matching parameters corresponding to the points to be matched in each level as a reliability parameter to evaluate each level. All levels are arranged in ascending order according to the reliability parameter to obtain a sorting table. The reliability levels in this application are based on this sorting table.

[0029] Figure 2 A schematic diagram illustrating the hierarchical division of buffer regions according to an embodiment of the present disclosure is shown. Figure 2 As shown, point O is the target feature point, which corresponds to multiple points A, B, and C in the reference road network data. Based on the distance between the points to be matched and the target feature point, the buffer zone is divided into multiple levels. The first level is a 2-meter circle, and the remaining levels increase sequentially at 0.5-meter intervals until the final distance threshold is reached. Figure 2 As shown, point A to be matched is located in the fourth level, point B in the third level, and point C in the first level. The reliability parameter of each level can be evaluated based on the mean of the topological similarity matching parameters of the points to be matched in each level, and multiple levels can be ranked.

[0030] According to some embodiments, step S103 includes: obtaining an initial matching point set, the initial matching point set including each target feature point among the plurality of target feature points and a matching point pair consisting of a matching reference feature point; and determining the initial frequency distribution based on the initial matching point set. This provides the basis for initial statistics used in subsequent iterative operations.

[0031] According to some embodiments, the initial matching point set is obtained based on the following steps: for each of the plurality of target feature points, a matching point in the buffer region that matches the target feature point is determined based on a preset threshold, and a matching point pair is formed with the target feature point; the matching point pair corresponding to each of the plurality of target feature points is obtained to obtain the initial matching point set. Thus, the initial matching point set can be obtained based on the above method, and this most reliable matching result can be used as the basis for initial statistics in subsequent iterative operations.

[0032] For example, a strict threshold can be used to filter the initial set of matching points. The topological similarity parameter needs to be greater than a certain threshold, and the angle difference needs to be less than a certain threshold, so that for each target feature point, there is only one reference feature point corresponding to it in the buffer area, and for that reference feature point, there is only one target feature point corresponding to it within the distance threshold range.

[0033] According to some embodiments, the initial frequency distribution can be determined by the following method: calculating the distance similarity matching parameter, angle similarity matching parameter, shape similarity matching parameter, and topological similarity matching parameter corresponding to each pair of matching points in the initial matching point set; and determining the frequency distribution corresponding to each of the four similarity matching parameters to determine the initial frequency distribution. Among these four similarity matching parameters, only the topological similarity matching parameter can directly express the probability between the point to be matched and the target feature point, while the distance similarity matching parameter, angle similarity matching parameter, and shape similarity matching parameter need to be mapped to obtain the matching probability. The matching probability between the target feature point and the point to be matched can be calculated based on the probabilities of the corresponding parameters obtained from the above mapping.

[0034] According to some embodiments, distance similarity matching parameters, angle similarity matching parameters, and shape similarity matching parameters can be mapped to probabilities using statistical methods. Specifically, the similarity matching parameters of the initial matching point set can first be statistically analyzed to obtain the frequency distribution map corresponding to each similarity matching parameter. Then, the similarity matching parameters of the points to be matched are input into the frequency distribution map to obtain the corresponding evaluation probabilities. In the process of statistically analyzing the frequency distribution map, the input similarity matching parameters are first sorted, and the frequency distribution of all similarity matching parameters is statistically analyzed by setting the number of value intervals.

[0035] In one example, during the process of mapping similarity matching parameters using a statistical method, the length of the value interval can be increased by gradually reducing the number of value intervals. When the frequency value of each interval is greater than 0, it is checked whether there is a situation where the frequency value of a certain interval is too small compared to the frequency value of the adjacent intervals. If so, the number of intervals is reduced until there are no excessively small frequency values, or until the set minimum number of intervals threshold is reached, so as to avoid the frequency distribution map having a value of 0 or too small in a certain interval, thereby improving the accuracy of the matching probability calculation.

[0036] For example, since each similarity feature parameter conforms to a certain frequency distribution—for instance, distance and angular similarity follow an exponential distribution, and shape similarity follows a normal distribution—the current distribution parameters can be obtained by statistically analyzing the initial set of matching points, thus deriving the frequency distribution formula. Input values ​​exceeding the current statistical range can have their probability values ​​estimated using the current distribution, allowing this method to have a wider statistical range.

[0037] According to some embodiments, step S104 includes: for each of the plurality of target feature points, determining the matching probability of the matching point corresponding to the highest reliability parameter level of the target feature point based on the initial frequency distribution to obtain an initial probability matrix, wherein the rows in the initial probability matrix represent the matching probability between one of the plurality of target feature points and the plurality of matching points corresponding to that target feature point; performing a filtering operation on the initial probability matrix to output a probability matrix, the filtering operation ensuring that in the output probability matrix, for each target feature point, there is at most one matching point that matches it; updating the initial matching point set based on the probability matrix to obtain a matching point set; updating the initial frequency distribution based on the matching point set; sorting the plurality of levels according to the reliability parameter from high to low to obtain a sorting table; and for each of the plurality of target feature points, based on the updated initial frequency distribution, obtaining the matching probability corresponding to the currently traversed level and the matching point in the next level in the sorting table, and outputting the updated probability matrix, until all levels have been traversed.

[0038] The probability matrix is ​​built iteratively to adaptively adjust it based on data. First, an initial probability matrix is ​​established based on the initial frequency distribution. Specifically, the initial frequency distribution can be used to evaluate the similarity matching parameters corresponding to the feature points to be matched in the highest reliability level of the ranking table. The similarity matching parameters of the point to be matched are input into the frequency distribution map to obtain probability estimates of these parameters. The probability estimates of each similarity matching parameter are then weighted to obtain the final matching probability between the point to be matched and the target feature point. This probability value is then input into the initial probability matrix for subsequent iterations. In the initial probability matrix, each row represents the matching probability between the target feature point in the target road network data and all point to be matched in the reference road network data, and each column represents the matching probability between a reference feature point in the reference road network data and the target feature point in the target road network data.

[0039] After obtaining the initial probability matrix, the optimal matching point pairs can be selected based on this matrix. Specifically, a marker matrix can be defined, which is a two-dimensional table with the same number of rows and columns as the initial probability matrix. The values ​​in the marker matrix include: -1 indicating that the point pair cannot match; 0 indicating that the point pair does not match; and 1 indicating that the point pair matches. The initial value of the marker matrix is ​​set to 0. First, the largest probability value in each row of the initial probability matrix is ​​selected and marked as 1. After completing the row filtering, conflicts may occur, for example, there may be multiple maximum values ​​in the j-th column, meaning that multiple points in the target road network data correspond to R in the parameter road network data. jIn this case, these columns need to be processed to eliminate conflicts. For columns with multiple row maximum values, first select the highest probability in that column and mark the non-maximum values ​​in that column as -1, meaning no further matching is possible. Since each target feature point has at most one match, for the rows marked as -1 in this filtering process, find the maximum probability of the match pair marked as 0 and mark it as 1. Then perform column checks, repeating this process until all conflicts are eliminated, outputting the probability matrix corresponding to the highest reliability level.

[0040] According to some embodiments, matching point pairs can be determined based on the probability matrix corresponding to the highest reliability parameter level, and the initial matching point set is updated to obtain the final matching point set. The frequency distribution of each similarity matching parameter in the current matching point set is statistically analyzed, and this frequency distribution is used to evaluate all currently traversed reliability parameter levels and the matching points for the next reliability parameter level. This process is iterated until all levels have been traversed, and the processed probability matrix is ​​output. The filtering operation on the probability matrix during each iteration is similar to the filtering operation on the initial probability matrix, and will not be described in detail here.

[0041] Figure 3 A schematic diagram of a probability matrix according to an embodiment of the present disclosure is shown. P ij Represents the target feature point T in the target road network data. i Reference feature point R in the reference road network data j The matching probability between them, where the i-th row represents the target feature point T in the target road network data. i The matching probability with the target feature point in the reference road network data. n represents the number of target feature points in the target road network data, and m represents the number of reference feature points in the reference road network data. If two points from the two sets of road network data are not the target points, then the probability of these two points is marked as 0.

[0042] According to some embodiments, for each of the plurality of target feature points, based on the distance similarity matching parameter frequency distribution in the initial frequency distribution, a first matching probability is determined for the point to be matched in the highest level of reliability parameter corresponding to the target feature point; based on the angle similarity matching parameter frequency distribution in the initial frequency distribution, a second matching probability is determined for the point to be matched in the highest level of reliability parameter corresponding to the target feature point; based on the shape similarity matching parameter frequency distribution in the initial frequency distribution, a third matching probability is determined for the point to be matched in the highest level of reliability parameter corresponding to the target feature point; and based on the first matching probability, the second matching probability, the third matching probability, and the topological similarity matching parameter corresponding to the point to be matched, the matching probability between the point to be matched and the target feature point is determined.

[0043] According to some embodiments, the matching probability between the target feature point and the point to be matched is obtained by fusing different similarity matching parameters together using a normalized weight formula. The matching probability p between target feature point i and point j to be matched is... i，j It can be calculated using the following formula:

[0044] p i，j =ε1p_Distance i，j +ε2p_Angle_similarity i，j +ε3p_Shape_similarity i，j +ε4p_Topo_similarity i，j

[0045] Where p_Distance is the distance similarity matching parameter between target feature point i and point j to be matched, p_Angle_similarity is the angle similarity matching parameter between target feature point i and point j to be matched, p_Shape_similarity is the shape similarity matching parameter between target feature point i and point j to be matched, and p_Topo_similarity is the topological similarity matching parameter between target feature point i and point j to be matched. ε1, ε2, ε3, and ε4 are weight parameters that are greater than zero, and ε1+ε2+ε3+ε4=1.

[0046] According to some embodiments, a compatibility factor corresponding to the probability matrix is ​​obtained, and the support parameter is calculated based on the compatibility factor, wherein the compatibility factor represents the compatibility of the adjacent matched point pairs of the point to be matched with the point to be matched; the probability matrix is ​​relaxed iteratively based on the support parameter until convergence. Thus, by relaxing the probability matrix based on contextual information, a road network matching result with a higher degree of global optimization is obtained.

[0047] According to some embodiments, the following formula is used to describe the support ST of the surrounding matched point pairs for the unmatched point pair (Ti, Rj). ij (r) :

[0048]

[0049] Where i and h represent the target feature points T in the target road network data. i and T h The index values ​​j and k represent reference feature points R in the reference road network data. j and R k The index value. C(i,j;h,k) represents the current matched point pair (T). h R k For the current pair of points to be matched (T) i R j The compatibility factor of ) . m is the target road network data for T i Contributing target feature point T h The number of points. Support(i) represents the set of points in the matching point set that contribute to the matching of point Ti, where r represents the current number of iterations.

[0050] For the compatibility factor C, the similarity between matching vectors can be used to describe the compatibility between surrounding matched points and the currently evaluated pair of points to be matched. Regarding the degree of similarity between vectors, the magnitude of the matching vector represents the degree of offset between the matched point pairs, and the angular difference between the vectors represents the rotation of the two current road network databases. Therefore, the length and angle need to be combined to obtain the compatibility factor as follows:

[0051]

[0052] Among them, V ij Represents the target feature point T in the target road network data. i To the reference feature point R in the reference road network data j The vector, V hk Represents the target feature point T in the target road network data. h To the reference feature point R in the reference road network data k The vector.

[0053] Therefore, the probability matrix can be iteratively updated using the following formula until the probability value p is reached in two consecutive iterations. ij If the change is less than a certain range, it indicates that the update process has converged.

[0054]

[0055] Within a certain spatial range, the two road network databases are within a certain range in terms of geographical offset, deformation, and rotation. Therefore, matching points that are far apart have a smaller impact on the current point to be matched. The support point set Support(i) needs to be set within a certain range. This disclosure sets the target feature points in the support point set Support(i) of the target road network data and the point T to be updated. i The distance is the average length of the road network in the current road network data. This setting not only improves the efficiency of the calculation but also its accuracy, resulting in a more globally optimized matching result.

[0056] According to another aspect of this disclosure, a road network matching device is provided. For example... Figure 4 As shown, the road network matching device 400 includes: a first acquisition module 401, configured to acquire target road network data and reference road network data, wherein the target road network data includes multiple target feature points, and the reference road network data includes multiple reference feature points; a first determination module 402, configured to, for each of the multiple target feature points, determine multiple matchable points corresponding to the target feature point among the multiple reference feature points, and divide the multiple matchable points into multiple levels; a second acquisition module 403, configured to acquire an initial frequency distribution, the initial frequency distribution being used to map multiple similarity matching parameters into probability values; and an iteration module 404, configured to, based on the initial frequency distribution, according to... The reliability parameters are arranged in descending order, and iterative operations are performed on the multiple levels to obtain an updated frequency distribution based on the probability matrix output by each iteration for the next iteration. Each iteration operates on the matching points in the already traversed levels and the levels to be traversed. A third acquisition module 405 is configured to acquire the probability matrix output by the iterative operation. A fourth acquisition module 406 is configured to acquire the support parameter corresponding to the probability matrix and perform relaxation iteration on the probability matrix based on the support parameter. A second determination module 407 is configured to determine the matching relationship between the reference road network data and the target road network data based on the probability matrix obtained from the relaxation iteration.

[0057] The first determining module 402 obtains the points to be matched based on the target road network data and the reference road network data, and classifies the points to be matched into levels. The iterative module 404 uses the initial frequency distribution to evaluate the matching probability of the points to be matched that have been traversed to the current level and the next level, iterating in this way until all levels have been traversed, thereby obtaining a preliminary matching probability matrix. Then, the fourth acquiring module 406 iteratively updates the probability matrix based on context information to obtain a globally optimized probability matrix, which is used to determine the matching relationship between the reference road network data and the target road network data. Thus, the probability matrix established based on the two sets of input road network data, and the road network data matching based on this probability matrix, can be gradually iterated and adjusted according to the input data to adapt to the statistical characteristics of the data, in order to obtain the best matching results, which can effectively improve the accuracy of road network matching.

[0058] The operation of modules 401-407 of the road network matching device 400 is similar to the operation of steps S101-S107 described above, and will not be repeated here.

[0059] According to some embodiments, the first determining module 402 includes: a first determining unit configured to, for each of the plurality of target feature points, create a circular buffer in the target road network data with the target feature point as the center and a preset length as the radius; a second determining unit configured to, based on the circular buffer, perform overlay analysis in the reference road network data to determine the buffer area corresponding to the target feature point in the reference road network data, wherein the buffer area is an estimated area where a reference feature point matching the target feature point is located; and a third determining unit configured to determine the plurality of points to be matched corresponding to the target feature point in the buffer area. Thus, the first determining module 402 uses a buffer area analysis method to obtain points to be matched based on the target road network data and the reference road network data. It can determine the size of the buffer area based on an empirical threshold to perform preliminary screening of reference feature points, while ensuring that the points to be matched fall within the buffer area. Further analysis of the buffer area accurately locates the reference feature point matching the target feature point.

[0060] According to some embodiments, the first determining module 402 further includes: a partitioning unit configured to divide the buffer region into multiple levels based on the distance from the center of the buffer region; and a fourth determining unit configured to determine the level corresponding to each of the multiple points to be matched. The device further includes: a calculation module configured to calculate, for each of the multiple levels, the mean of the topological similarity matching parameters corresponding to the points to be matched in that level as a reliability parameter corresponding to that level; and a sorting module configured to sort the multiple levels based on the reliability parameter.

[0061] It is understandable that distance is one of the parameters for measuring whether two points match, and intuitively, the closer two points are, the higher the probability of a match. However, the inventors discovered that in the entire road network data, the probability of two points matching is not necessarily higher the closer they are. Therefore, the third determining module 404 uses the mean value of the topological similarity matching parameters corresponding to the points to be matched in each level as a reliability parameter to evaluate each level. All levels are sorted in ascending order according to the reliability parameters to obtain a sorting table. The reliability of the levels in this application is based on this sorting table.

[0062] According to some embodiments, the second acquisition module 403 includes: a first acquisition unit configured to acquire an initial matching point set, the initial matching point set including each of the plurality of target feature points and a matching point pair consisting of each target feature point and a matching reference feature point; and a fifth determination unit configured to determine the initial frequency distribution based on the initial matching point set. This provides the basis for initial statistics to be used in subsequent iterative operations.

[0063] According to some embodiments, the first acquisition unit includes: a sixth determining unit, configured to determine, for each of the plurality of target feature points, a matching point in the buffer region that matches the target feature point based on a preset threshold, and form a matching point pair with the target feature point; and a second acquisition unit, configured to acquire the matching point pair corresponding to each of the plurality of target feature points to obtain the initial matching point set.

[0064] For example, the first acquisition unit can use strict thresholds to filter the initial matching point set. The topological similarity parameter needs to be greater than a certain threshold, and the angle difference needs to be less than a certain threshold, so that for each target feature point, there is only one reference feature point corresponding to it in the buffer area, and for that reference feature point, there is only one target feature point corresponding to it within the distance threshold range.

[0065] According to some embodiments, the fifth determining unit includes: a calculation unit configured to calculate the distance similarity matching parameter, angle similarity matching parameter, shape similarity matching parameter, and topological similarity matching parameter corresponding to each pair of matching points in the initial matching point set; and a seventh determining unit configured to determine the frequency distribution corresponding to the similarity matching parameter, angle similarity matching parameter, and shape similarity matching parameter based on the distance similarity matching parameter, angle similarity matching parameter, and shape similarity matching parameter corresponding to each pair of matching points to determine the initial frequency distribution.

[0066] According to some embodiments, the calculation unit can map distance similarity matching parameters, angle similarity matching parameters, and shape similarity matching parameters into probabilities using statistical methods. Specifically, the calculation unit can first statistically analyze the similarity matching parameters of the initial matching point set to obtain the frequency distribution map corresponding to each similarity matching parameter. Then, the similarity matching parameters of the points to be matched are input into the frequency distribution map to obtain the corresponding evaluation probabilities. In the process of statistically analyzing the frequency distribution map, the input similarity matching parameters are first sorted, and the frequency distribution of all similarity matching parameters is statistically analyzed by setting the number of value intervals.

[0067] In one example, during the process of mapping similarity matching parameters using statistical methods, the computing unit can increase the length of the value intervals by gradually reducing the number of value intervals. When the frequency value of each interval is greater than 0, it checks whether there is a situation where the frequency value of a certain interval is too small compared to the frequency value of the adjacent intervals. If so, it continues to reduce the number of intervals until there are no excessively small frequency values, or until the set minimum number of intervals threshold is reached, in order to avoid the frequency distribution map having a value of 0 or too small in a certain interval, thereby improving the accuracy of the matching probability calculation.

[0068] For example, since each similarity feature parameter conforms to a certain frequency distribution—for instance, distance and angular similarity conform to an exponential distribution, and shape similarity conforms to a normal distribution—the computational unit can also obtain the current distribution parameters by statistically analyzing the initial set of matching points, thereby deriving the frequency distribution formula. Input values ​​exceeding the current statistical range can have their probability values ​​estimated using the current distribution, thus enabling this method to have a wider statistical range.

[0069] According to some embodiments, the iteration module 404 includes: an eighth determining unit, configured to, for each of the plurality of target feature points, determine the matching probability corresponding to the matching point in the highest level of reliability parameter corresponding to the target feature point based on the initial frequency distribution, to obtain an initial probability matrix, wherein the rows in the initial probability matrix represent the matching probability between one of the plurality of target feature points and the plurality of matching points corresponding to that target feature point; and a filtering unit, configured to perform a filtering operation on the initial probability matrix to output a probability matrix, wherein the filtering operation ensures that, in the output probability matrix, for each target feature point... At most one match point is found; a first update unit is configured to update the initial matching point set based on the probability matrix to obtain a matching point set; a second update unit is configured to update the initial frequency distribution based on the matching point set; a sorting unit is configured to sort the multiple levels in descending order of the reliability parameter to obtain a sorting table; and an output unit is configured to, for each of the multiple target feature points, obtain the matching probability corresponding to the currently traversed level and the match point in the next level in the sorting table based on the updated initial frequency distribution, and output the updated probability matrix, until all levels have been traversed.

[0070] The probability matrix is ​​built iteratively to adaptively adjust it based on data. First, the eighth determining unit needs to establish an initial probability matrix based on the initial frequency distribution. Specifically, the eighth determining unit can use the initial frequency distribution to evaluate the similarity matching parameters corresponding to the feature points to be matched in the highest reliability level of the ranking table. The similarity matching parameters of the points to be matched are input into the frequency distribution map to obtain probability estimates of these parameters. The probability estimates of each similarity matching parameter are then weighted to obtain the final matching probability between the point to be matched and the target feature point. This probability value is then input into the initial probability matrix for subsequent iterations. In the initial probability matrix, each row represents the matching probability between the target feature point in the target road network data and all points to be matched in the reference road network data, and each column represents the matching probability between a reference feature point in the reference road network data and the target feature point in the target road network data.

[0071] After obtaining the initial probability matrix, the filtering unit can select the optimal matching point pairs based on this matrix. Specifically, a marker matrix can be defined. The marker matrix is ​​a two-dimensional table with the same number of rows and columns as the initial probability matrix. The values ​​in the marker matrix include: -1 indicating that the point pair cannot match; 0 indicating that the point pair does not match; and 1 indicating that the point pair matches. The initial value of the marker matrix is ​​set to 0. First, the largest probability value in each row of the initial probability matrix is ​​selected and marked as 1. After completing the row filtering, conflicts may occur. For example, there may be multiple maximum values ​​in the j-th column, meaning that multiple points in the target road network data correspond to R values ​​in the parameter road network data. j In this case, these columns need to be processed to eliminate conflicts. For columns with multiple row maximum values, first select the highest probability in that column and mark the non-maximum values ​​in that column as -1, meaning no further matching is possible. Since each target feature point has at most one match, for the rows marked as -1 in this filtering process, find the maximum probability of the match pair marked as 0 and mark it as 1. Then perform column checks, repeating this process until all conflicts are eliminated, outputting the probability matrix corresponding to the highest reliability level.

[0072] According to some embodiments, the first update unit can determine matching point pairs based on the probability matrix corresponding to the highest reliability parameter level, and update the initial matching point set to obtain the matching point set. The second update unit statistically analyzes the frequency distribution of each similarity matching parameter in the current matching point set, and uses the current frequency distribution to evaluate all currently traversed reliability parameter levels and the matching points to be matched at the next reliability parameter level. The above process is iterated until all levels have been traversed, and the output unit outputs the processed probability matrix.

[0073] According to some embodiments, the eighth determining unit includes: a ninth determining unit configured to, for each of the plurality of target feature points, determine a first matching probability corresponding to a point to be matched in the highest level of reliability parameters corresponding to the target feature point, based on the frequency distribution of distance similarity matching parameters in the initial frequency distribution; a tenth determining unit configured to determine a second matching probability corresponding to a point to be matched in the highest level of reliability parameters corresponding to the target feature point, based on the frequency distribution of angle similarity matching parameters in the initial frequency distribution; an eleventh determining unit configured to determine a third matching probability corresponding to a point to be matched in the highest level of reliability parameters corresponding to the target feature point, based on the frequency distribution of shape similarity matching parameters in the initial frequency distribution; and a twelfth determining unit configured to determine a matching probability between the point to be matched and the target feature point based on the first matching probability, the second matching probability, the third matching probability, and the topological similarity matching parameter corresponding to the point to be matched.

[0074] According to some embodiments, the matching probability between the target feature point and the point to be matched is obtained by the twelfth determining unit by fusing different similarity matching parameters together using a normalized weight formula. The matching probability p between target feature point i and point j to be matched is... i,j It can be calculated using the following formula:

[0075] p i，j =ε1p_Distance i，j +ε2p_Angle_similarity i，j +ε3p_Shape_similarity i，j

[0076] +ε4p_Topo_similarity i，j

[0077] Where p_Distance is the distance similarity matching parameter between target feature point i and point j to be matched, p_Angle_similarity is the angle similarity matching parameter between target feature point i and point j to be matched, p_Shape_similarity is the shape similarity matching parameter between target feature point i and point j to be matched, and p_Topo_similarity is the topological similarity matching parameter between target feature point i and point j to be matched. ε1, ε2, ε3, and ε4 are weight parameters that are greater than zero, and ε1+ε2+ε3+ε4=1.

[0078] According to some embodiments, the second determining module 407 includes: a third acquisition unit, configured to acquire a compatibility factor corresponding to the probability matrix, and calculate the support parameter based on the compatibility factor, wherein the compatibility factor represents the compatibility of the adjacent matched point pairs of the point to be matched with the point to be matched; and an iteration unit, configured to perform relaxation iteration on the probability matrix based on the support parameter until convergence.

[0079] According to some embodiments, the third acquisition unit uses the following formula to describe the support ST of the surrounding matched point pairs for the unmatched point pair (Ti,Rj). ij (r) :

[0080]

[0081] Where i and h represent the target feature points T in the target road network data. i and T h The index values ​​j and k represent reference feature points R in the reference road network data. j and R kThe index value. C(i,j; h,k) represents the current matched point pair (T). h ,R k For the current pair of points to be matched (T) i ,R j The compatibility factor of ) . m is the target road network data for T i Contributing target feature point T h The number of points. Support(i) represents the set of points in the matching point set that contribute to the matching of point Ti, where r represents the current number of iterations.

[0082] For the compatibility factor C, the third acquisition unit can use the similarity between matching vectors to describe the compatibility between surrounding matched points and the currently evaluated pair of points to be matched. Regarding the degree of similarity between vectors, the magnitude of the matching vector represents the degree of offset between the matched point pairs, and the angular difference between the vectors represents the rotation of the two current road network databases. Therefore, the length and angle need to be combined to obtain the compatibility factor as follows:

[0083]

[0084] Among them, V ij Represents the target feature point T in the target road network data. i To the reference feature point R in the reference road network data j The vector, V hk Represents the target feature point T in the target road network data. h To the reference feature point R in the reference road network data k The vector.

[0085] Therefore, the probability matrix can be iteratively updated using the following formula until the probability value p is reached in two consecutive iterations. ij If the change is less than a certain range, it indicates that the update process has converged.

[0086]

[0087] Within a certain spatial range, the two road network databases are within a certain range in terms of geographical offset, deformation, and rotation. Therefore, matching points that are far apart have a smaller impact on the current point to be matched. The support point set Support(i) needs to be set within a certain range. This disclosure sets the target feature points in the support point set Support(i) of the target road network data and the point T to be updated. i The distance is the average length of the road network in the current road network data. This setting not only improves the efficiency of the calculation but also its accuracy, resulting in a more globally optimized matching result.

[0088] According to embodiments of this disclosure, an electronic device, a readable storage medium, and a computer program product are also provided.

[0089] refer to Figure 5 The present invention describes a structural block diagram of an electronic device 500 that can serve as a server or client of the present disclosure, which is an example of a hardware device that can be applied to various aspects of the present disclosure. The electronic device is intended to represent various forms of digital electronic computer devices, such as 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 personal digital processors, cellular phones, smartphones, wearable devices, and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely illustrative and are not intended to limit the implementation of the present disclosure described and / or claimed herein.

[0090] like Figure 5 As shown, the electronic device 500 includes a computing unit 501, which can perform various appropriate actions and processes according to a computer program stored in a read-only memory (ROM) 502 or a computer program loaded from a storage unit 508 into a random access memory (RAM) 503. The RAM 503 may also store various programs and data required for the operation of the electronic device 500. The computing unit 501, ROM 502, and RAM 503 are interconnected via a bus 504. An input / output (I / O) interface 505 is also connected to the bus 504.

[0091] Multiple components in electronic device 500 are connected to I / O interface 505, including: input unit 506, output unit 507, storage unit 508, and communication unit 509. Input unit 506 can be any type of device capable of inputting information to electronic device 500. Input unit 506 can receive input digital or character information and generate key signal inputs related to user settings and / or function control of electronic device, and may include, but is not limited to, a mouse, keyboard, touchscreen, trackpad, trackball, joystick, microphone, and / or remote control. Output unit 507 can be any type of device capable of presenting information, and may include, but is not limited to, a monitor, speaker, video / audio output terminal, vibrator, and / or printer. Storage unit 508 may include, but is not limited to, disk and optical disk. Communication unit 509 allows electronic device 500 to exchange information / data with other devices through computer networks such as the Internet and / or various telecommunications networks, and may include, but is not limited to, modems, network cards, infrared communication devices, wireless communication transceivers, and / or chipsets, such as Bluetooth™ devices, 802.11 devices, WiFi devices, WiMax devices, cellular communication devices, and / or the like.

[0092] The computing unit 501 can be a variety of general-purpose and / or special-purpose processing components with processing and computing capabilities. Some examples of the computing unit 501 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various special-purpose 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 501 performs the various methods and processes described above, such as road network matching methods. For example, in some embodiments, the road network matching method may be implemented as a computer software program tangibly contained in a machine-readable medium, such as storage unit 508. In some embodiments, part or all of the computer program may be loaded and / or installed on the electronic device 500 via ROM 502 and / or communication unit 509. When the computer program is loaded into RAM 503 and executed by the computing unit 501, one or more steps of the road network matching method described above may be performed. Alternatively, in other embodiments, the computing unit 501 may be configured to perform the road network matching method by any other suitable means (e.g., by means of firmware).

[0093] Various embodiments of the systems and techniques described above herein can be implemented in digital electronic circuit systems, integrated circuit systems, field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), systems-on-a-chip (SoCs), complex programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments may include implementations in one or more computer programs that can be executed and / or interpreted on a programmable system including at least one programmable processor, which may be a dedicated or general-purpose programmable processor, capable of receiving data and instructions from a storage system, at least one input device, and at least one output device, and transmitting data and instructions to the storage system, the at least one input device, and the at least one output device.

[0094] The program code used to implement the methods of this disclosure may be written in any combination of one or more programming languages. This program code may be provided to a processor or controller of a general-purpose computer, special-purpose computer, or other programmable data processing apparatus, such that when executed by the processor or controller, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be implemented. The program code may be executed entirely on a machine, partially on a machine, as a standalone software package partially on a machine and partially on a remote machine, or entirely on a remote machine or server.

[0095] In the context of the present disclosure, a machine-readable medium can be a tangible medium that can contain or store a program for use by or in conjunction with an instruction execution system, device or equipment. 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, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or equipment, or any suitable combination of the foregoing. A more specific example of a machine-readable storage medium can include an electrical connection based on one or more lines, a portable computer disk, 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 disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing.

[0096] 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 pointing device (e.g., a mouse or trackball) through which the user can provide input to the computer. Other types 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).

[0097] The systems and technologies described herein can be implemented in computing systems that include backend components (e.g., as a data server), or computing systems that include middleware components (e.g., an application server), or computing systems that include frontend components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with embodiments of the systems and technologies described herein), or any combination of such backend, middleware, or frontend components. The components of the system can be interconnected via digital data communication of any form or medium (e.g., a communication network). Examples of communication networks include local area networks (LANs), wide area networks (WANs), and the Internet.

[0098] Computer systems can include clients and servers. Clients and servers are generally located far apart and typically interact via communication networks. Client-server relationships are created by computer programs running on the respective computers and having a client-server relationship with each other. Servers can be cloud servers, servers in distributed systems, or servers incorporating blockchain technology.

[0099] It should be understood that the various forms of processes shown above can be used to rearrange, add, or delete steps. For example, the steps described in this disclosure can be performed in parallel, sequentially, or in a different order, as long as the desired result of the technical solution disclosed in this disclosure can be achieved, and this is not limited herein.

[0100] While embodiments or examples of this disclosure have been described with reference to the accompanying drawings, it should be understood that the methods, systems, and devices described above are merely exemplary embodiments or examples, and the scope of the invention is not limited by these embodiments or examples, but only by the granted claims and their equivalents. Various elements in the embodiments or examples may be omitted or replaced by their equivalents. Furthermore, the steps may be performed in a different order than that described in this disclosure. Further, various elements in the embodiments or examples may be combined in various ways. Importantly, as the technology evolves, many elements described herein can be replaced by equivalents that appear after this disclosure.

Claims

1. A road network matching method, comprising: Acquire target road network data and reference road network data, wherein the target road network data includes multiple target feature points and the reference road network data includes multiple reference feature points; For each of the plurality of target feature points, determine a plurality of match points corresponding to the target feature point among the plurality of reference feature points, and divide the plurality of match points into multiple levels; For each of the multiple levels, the mean value of the topological similarity matching parameters corresponding to the points to be matched in that level is calculated as the reliability parameter corresponding to that level. The multiple levels are sorted based on the reliability parameters; Obtain an initial frequency distribution, which is used to map multiple similarity matching parameters into probability values; Based on the initial frequency distribution, iterative operations are performed on the multiple levels to obtain an updated frequency distribution based on the probability matrix output by each iteration for the next iteration. Each iteration operates on the matching points in both the already traversed levels and the levels to be traversed. The iterative operations on the multiple levels based on the initial frequency distribution include: For each of the plurality of target feature points, the matching probability of the matching point corresponding to the highest level of reliability parameter of the target feature point is determined based on the initial frequency distribution to obtain an initial probability matrix, wherein the rows in the initial probability matrix represent the matching probability between one of the plurality of target feature points and the plurality of matching points corresponding to the target feature point; The initial probability matrix is ​​filtered to output a probability matrix, wherein in the output probability matrix, for each target feature point, there is at most one match point. The initial set of matching points is updated based on the probability matrix to obtain the set of matching points; Update the initial frequency distribution based on the set of matching points; The multiple levels are sorted in descending order of the reliability parameters to obtain a sorting table; and For each of the multiple target feature points, based on the updated initial frequency distribution, the matching probability corresponding to the currently traversed level and the matching point in the next level in the sorting table is obtained, and the updated probability matrix is ​​output until all levels have been traversed. Obtain the probability matrix output based on the iterative operation; Obtain the support parameters corresponding to the probability matrix, and perform relaxation iteration on the probability matrix based on the support parameters; and Based on the probability matrix obtained by relaxation iteration, the matching relationship between the reference road network data and the target road network data is determined.

2. The method according to claim 1, wherein, For each of the plurality of target feature points, determining the plurality of matchable points corresponding to that target feature point from the plurality of reference feature points includes: For each of the plurality of target feature points, a circular buffer zone is formed in the target road network data with the target feature point as the center and a preset length as the radius; Based on the circular buffer, overlay analysis is performed on the reference road network data to determine the buffer region corresponding to the target feature point in the reference road network data, wherein the buffer region is the estimated region where the reference feature point matching the target feature point is located; and Determine the plurality of points to be matched in the buffer region that correspond to the target feature point.

3. The method according to claim 2, wherein, The step of dividing the plurality of points to be matched into multiple levels includes: Based on the distance from the center of the buffer area, the buffer area is divided into multiple levels; Determine the level corresponding to each of the plurality of points to be matched.

4. The method according to claim 2, wherein, The process of obtaining the initial frequency distribution includes: Obtain an initial matching point set, which includes a matching point pair consisting of each of the plurality of target feature points and a matching reference feature point; The initial frequency distribution is determined based on the initial set of matching points.

5. The method according to claim 4, wherein, The initial set of matching points is obtained based on the following steps: For each of the plurality of target feature points, a matching point in the buffer region that matches the target feature point is determined based on a preset threshold, and a matching point pair is formed with the target feature point; Obtain the matching point pair corresponding to each of the plurality of target feature points to obtain the initial matching point set.

6. The method according to claim 4, wherein, Determining the initial frequency distribution based on the initial matching point set includes: Calculate the distance similarity matching parameter, angle similarity matching parameter, shape similarity matching parameter, and topological similarity matching parameter for each pair of matching points in the initial matching point set; Based on the distance similarity matching parameters, angle similarity matching parameters, and shape similarity matching parameters corresponding to each matching point pair, the frequency distributions corresponding to the similarity matching parameters, angle similarity matching parameters, and shape similarity matching parameters are determined to determine the initial frequency distribution.

7. The method according to claim 1, wherein, The step of determining the matching probability of the matching point corresponding to the highest level of reliability parameter among the plurality of target feature points based on the initial frequency distribution for each target feature point includes: For each of the plurality of target feature points, Based on the frequency distribution of distance similarity matching parameters in the initial frequency distribution, the first matching probability corresponding to the matching point in the highest level of reliability parameters corresponding to the target feature point is determined; Based on the frequency distribution of angle similarity matching parameters in the initial frequency distribution, the second matching probability corresponding to the matching point in the highest level of reliability parameters corresponding to the target feature point is determined; Based on the shape similarity matching parameter frequency distribution in the initial frequency distribution, determine the third matching probability corresponding to the matching point in the highest level of reliability parameter corresponding to the target feature point; and Based on the first matching probability, the second matching probability, the third matching probability, and the topological similarity matching parameter corresponding to the point to be matched, the matching probability between the point to be matched and the target feature point is determined.

8. The method according to any one of claims 1 to 7, wherein, The step of obtaining the support parameters corresponding to the probability matrix and performing relaxation iteration on the probability matrix based on the support parameters includes: Obtain the compatibility factor corresponding to the probability matrix, and calculate the support parameter based on the compatibility factor, wherein the compatibility factor represents the compatibility of the adjacent matched point pairs of the point to be matched with the point to be matched; The probability matrix is ​​relaxed iterated based on the support parameter until convergence.

9. A road network matching device, comprising: The first acquisition module is configured to acquire target road network data and reference road network data, wherein the target road network data includes multiple target feature points and the reference road network data includes multiple reference feature points; The first determining module is configured to, for each of the plurality of target feature points, determine a plurality of match points corresponding to the target feature point among the plurality of reference feature points, and divide the plurality of match points into a plurality of levels; The calculation module is configured to calculate the mean of the topological similarity matching parameters corresponding to the points to be matched in each of the plurality of levels, and use it as the reliability parameter corresponding to that level. The sorting module is configured to sort the multiple levels based on the reliability parameters; the second acquisition module is configured to acquire an initial frequency distribution, which is used to map multiple similarity matching parameters into probability values. An iteration module is configured to perform iterative operations on the multiple levels based on the initial frequency distribution, to obtain an updated frequency distribution based on the probability matrix output by each iteration for the next iteration. Each iteration operation targets the matching points in the already traversed levels and the levels to be traversed. The iteration module includes: The eighth determining unit is configured to, for each of the plurality of target feature points, determine the matching probability of the matching point corresponding to the highest level of reliability parameter of the target feature point based on the initial frequency distribution, so as to obtain an initial probability matrix, wherein the rows in the initial probability matrix represent the matching probability between one of the plurality of target feature points and the plurality of matching points corresponding to the target feature point; The filtering unit is configured to perform a filtering operation on the initial probability matrix to output a probability matrix, the filtering operation being such that in the output probability matrix, for each target feature point, there is at most one match point. The first update unit is configured to update the initial matching point set based on the probability matrix to obtain the matching point set; The second update unit is configured to update the initial frequency distribution based on the set of matching points; A sorting unit is configured to sort the plurality of levels in descending order of the reliability parameters to obtain a sorting table; and The output unit is configured to, for each of the plurality of target feature points, obtain the matching probability corresponding to the currently traversed level and the matching point in the next level in the sorting table based on the updated initial frequency distribution, and output the updated probability matrix until all levels have been traversed. The third acquisition module is configured to acquire the probability matrix output based on the iterative operation; The fourth acquisition module is configured to acquire the support parameters corresponding to the probability matrix, and perform relaxation iteration on the probability matrix based on the support parameters; and The second determining module is configured to determine the matching relationship between the reference road network data and the target road network data based on the probability matrix obtained by relaxation iteration.

10. The apparatus according to claim 9, wherein, The first determining module includes: The first determining unit is configured to, for each of the plurality of target feature points, create a circular buffer in the target road network data with the target feature point as the center and a preset length as the radius; The second determining unit is configured to perform overlay analysis on the reference road network data based on the circular buffer to determine the buffer region corresponding to the target feature point in the reference road network data, wherein the buffer region is an estimated region where a reference feature point matching the target feature point is located; and The third determining unit is configured to determine the plurality of points to be matched in the buffer region corresponding to the target feature point.

11. The apparatus according to claim 10, wherein, The first determining module further includes: The partitioning unit is configured to divide the buffer region into multiple levels based on the distance from the center of the buffer region. The fourth determining unit is configured to determine the level corresponding to each of the plurality of points to be matched.

12. The apparatus according to claim 10, wherein, The second acquisition module includes: The first acquisition unit is configured to acquire an initial matching point set, wherein the initial matching point set includes a matching point pair consisting of each of the plurality of target feature points and a matching reference feature point; The fifth determining unit is configured to determine the initial frequency distribution based on the initial matching point set.

13. The apparatus according to claim 12, wherein, The first acquisition unit includes: The sixth determining unit is configured to, for each of the plurality of target feature points, determine a matching point in the buffer region that matches the target feature point based on a preset threshold, and form a matching point pair with the target feature point; The second acquisition unit is configured to acquire the matching point pair corresponding to each of the plurality of target feature points to obtain the initial matching point set.

14. The apparatus according to claim 12, wherein, The fifth determining unit includes: The calculation unit is configured to calculate the distance similarity matching parameter, angle similarity matching parameter, shape similarity matching parameter and topological similarity matching parameter corresponding to each pair of matching points in the initial matching point set; The seventh determining unit is configured to determine the frequency distribution corresponding to the similarity matching parameters, angle similarity matching parameters, and shape similarity matching parameters for each matching point pair to determine the initial frequency distribution.

15. The apparatus according to claim 9, wherein, The eighth determining unit includes: The ninth determining unit is configured to, for each of the plurality of target feature points, determine the first matching probability corresponding to the point to be matched in the highest level of reliability parameter corresponding to the target feature point based on the frequency distribution of distance similarity matching parameters in the initial frequency distribution; The tenth determining unit is configured to determine the second matching probability corresponding to the matching point in the highest level of reliability parameter corresponding to the target feature point based on the frequency distribution of the angle similarity matching parameter in the initial frequency distribution; The eleventh determining unit is configured to determine, based on the shape similarity matching parameter frequency distribution in the initial frequency distribution, the third matching probability corresponding to the matching point in the highest level of reliability parameter corresponding to the target feature point; and The twelfth determining unit is configured to determine the matching probability between the point to be matched and the target feature point based on the first matching probability, the second matching probability, the third matching probability and the topological similarity matching parameter corresponding to the point to be matched.

16. The apparatus according to any one of claims 9 to 15, wherein, The second determining module includes: The third acquisition unit is configured to acquire the compatibility factor corresponding to the probability matrix and calculate the support parameter based on the compatibility factor, wherein the compatibility factor represents the compatibility of the adjacent matched point pairs of the point to be matched with the point to be matched. The iterative unit is configured to perform a relaxation iteration on the probability matrix based on the support parameter until convergence.

17. An electronic device comprising: At least one processor; as well as a memory communicatively coupled to the at least one processor; in The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method of any one of claims 1-8.

18. A non-transitory computer-readable storage medium storing computer instructions, wherein, The computer instructions are used to cause the computer to perform the method according to any one of claims 1-8.

19. A computer program product comprising a computer program, wherein, When the computer program is executed by a processor, it implements the method of any one of claims 1-8.

Images