Annular radial road network mode identification method, device and system, and storage medium

Abstract

The invention discloses an annular radial road network pattern recognition method, device and system, and a storage medium, and belongs to the technical field of urban traffic systems, and the method comprises the steps: S1, building a road network dual model; s2, extracting a skeleton line of the road network according to the road network dual model; s3, extracting a fuzzy radiation center, a radiation edge and a loop of the road network according to the skeleton line; and S4, evaluating the importance scores of the radiation edge and the loop based on the weight so as to realize annular radiation mode discrimination. By adopting the technical scheme of the invention, the pattern recognition efficiency and accuracy can be improved, and the structural characteristics of the road network can be better revealed.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban traffic systems, and in particular relates to a method and device, a system, and a storage medium for identifying a ring-shaped radial road network pattern. Background Art

[0002] As an important part of the urban transportation system, the road network plays a vital role in urban development. The road network pattern contains the mechanism of urban formation and evolution and reflects the layout characteristics of urban roads. Extracting and identifying the pattern characteristics of the road network has practical value for urban development, traffic management and traffic planning. Using geometric and topological methods to identify road network patterns has always been one of the research hotspots in map synthesis. Common road network patterns include ring-radiation pattern, grid pattern and irregular pattern. Road networks of different patterns are different in terms of connectivity, continuity and traffic efficiency. Therefore, the geometric and topological characteristics expressed by different types of road entities are different. The ring-radiation pattern road network usually contains a radiation center, from which the radiation center leads to radiation edges and loops; the grid pattern road network is generally composed of several mutually orthogonal roads; the irregular pattern road network has no obvious structure or regular shape. Road network pattern recognition is affected by the subject's psychological cognitive factors. Due to the complexity of the road network structure and the uncertainty of human cognition, road network pattern recognition is very difficult.

[0003] A ring-shaped radial road network refers to a road network in which a ring road forms the main skeleton of the road network, and there are multiple roads radially connected to the ring road. The ring-shaped radial road network has a small nonlinear coefficient, high accessibility, and good network formation rate. It is an ideal road network form, which is conducive to urban expansion and transit traffic diversion. There are few studies on the pattern recognition of ring-shaped radial road networks. Reference 1 (Tian Jing, Zhang Boyu, Wu Dang, A new method for radial pattern recognition of vector road networks, Journal of Wuhan University (Information Science Edition), 2013, 38(10): 1234-1238.) established a quantitative calculation model and expression model based on qualitative features and prior knowledge, and realized the radial pattern recognition at a macro scale. Although this method performs well in specific data sets or regions, its applicability to other complex road network environments needs further verification. Reference 2 (Cui Xiaojie, Research on the recognition method of typical spatial distribution patterns of urban road networks, 2019.) proposed a step-by-step recognition method based on graph theory for the ring-shaped radial pattern composed of urban ring roads and radial roads. Due to the step-by-step recognition strategy, multiple steps and a large number of parameter adjustments are involved, especially when identifying complex radiation patterns and ring patterns. This complexity not only increases the computational cost, but may also lead to uncertainty in the results. Reference 3 (HEINZLE F, ANDERS K H. Characterizing Space via Pattern Recognition Techniques: Identifying Patterns in Road Networks. Generalisation of geographic information. Elsevier Science BV. 2007: 233-253.) identifies urban centers through a clustering algorithm and uses Hough transform based on elliptic equations to detect ring structures. This method is mainly suitable for the recognition of simple radiation patterns, and is not effective for road networks with more complex ring-shaped radiation patterns. In summary, the existing methods perform well when dealing with simple patterns, but their applicability is limited for more complex urban mixed patterns or multi-center structures. In addition, methods for identifying complex patterns in road networks usually involve complex computational processes and are not efficient when dealing with large-scale urban networks. Summary of the invention

[0004] The technical problem to be solved by the present invention is to provide a method and device, system and storage medium for identifying a circular radial road network pattern. To achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for identifying a circular radial road network pattern, comprising:

[0006] Step S1, establishing a road network dual model; wherein the road network dual model is used to visualize the topological relationship of the original road network;

[0007] Step S2: According to the road network dual model, the importance of urban roads is evaluated using length, connectivity, betweenness and proximity as measurement indicators, and the skeleton line of the road network is extracted;

[0008] Step S3, extracting the fuzzy radiation center, radiation edge and loop of the road network according to the skeleton line; wherein, according to the skeleton line, the fuzzy radiation center is extracted by a hot spot analysis method; according to the skeleton line and the fuzzy radiation center, the radiation edge is extracted by a breadth-first search method and linearized; according to the skeleton line, the fuzzy radiation center and the radiation edge, the loop is extracted based on the angle and turn of the connecting road section;

[0009] Step S4: Evaluate the importance scores of the radiating edges and loops based on the weights to achieve ring-shaped radiation pattern discrimination.

[0010] The present invention also provides a ring-shaped radial road network pattern recognition device, comprising:

[0011] Establishing a module for establishing a road network dual model; wherein the road network dual model is used to visualize the topological relationship of the original road network;

[0012] The first extraction module is used to evaluate the importance of urban roads based on the road network dual model using length, connectivity, betweenness and proximity as measurement indicators, and extract the skeleton line of the road network;

[0013] The second extraction module is used to extract the fuzzy radiation center, radiation edge and loop of the road network according to the skeleton line; wherein the second extraction module extracts the fuzzy radiation center by a hot spot analysis method according to the skeleton line; the second extraction module extracts the radiation edge by a breadth-first search method according to the skeleton line and the fuzzy radiation center and performs linearization processing; the second extraction module extracts the loop according to the skeleton line, the fuzzy radiation center and the radiation edge based on the angle and turn of the connecting road section;

[0014] The identification module is used to evaluate the importance scores of the radiation edges and loops based on the weights to achieve ring radiation pattern discrimination.

[0015] An embodiment of the present invention also provides a ring-shaped radial road network pattern recognition system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a ring-shaped radial road network pattern recognition method when executed by the processor.

[0016] An embodiment of the present invention further provides a storage medium, on which a computer program is stored, and the computer program executes a method for identifying a ring-shaped radial road network pattern when running.

[0017] The present invention recognizes road network patterns through skeleton lines and identifies the annular radiation patterns of the road network by calculating relevant indicators, which can improve the efficiency and accuracy of pattern recognition and better reveal the structural characteristics of the road network. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying creative work.

[0019] Figure 1 It is a flow chart of a method for identifying a ring-shaped radial road network pattern according to an embodiment of the present invention;

[0020] Figure 2 Schematic diagram of the extraction of the radiation center; (a) is the road density of the mesh, and (b) is the POI density of the mesh;

[0021] Figure 3 Schematic diagram of radial edge extraction, where (a) is the search for candidate radial edges, (b) is the marking of candidate radial edges, (c) is the bifurcation check, and (d) is the selection of roads;

[0022] Figure 4 is a schematic diagram of the straightening of the radiation edge;

[0023] Figure 5 Schematic diagram of road section turning in the loop; (a) is a road section with consistent turning, and (b) is a road section with inconsistent turning;

[0024] Figure 6 Schematic diagram of the characteristic parameters of the loop. DETAILED DESCRIPTION

[0025] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0026] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.

[0027] Embodiment 1:

[0028] like Figure 1 As shown, an embodiment of the present invention provides a method for identifying a circular radial road network pattern, which utilizes the characteristics of skeleton lines while maintaining the main structural features of the road network and reducing the complexity of the road network spatial structure, and realizes effective identification of the road network pattern by analyzing the radial edges and loops; the method comprises the following steps:

[0029] Step S1, establishing a road network dual model;

[0030] Step S2, extract the skeleton line of the road network according to the road network dual model; Step S3, extract the fuzzy radiation center, radiation edge and loop of the road network according to the skeleton line; Step S4, evaluate the importance score of the radiation edge and loop based on the weight to realize the ring radiation pattern discrimination. As an implementation method of an embodiment of the present invention, in step S1, the road network dual model is used to visualize the topological relationship of the original road network, firstly generate the stroke of the road, and then establish the road network dual model based on the topological relationship of the stroke; wherein, in complex network theory, the dual graph refers to a transformation of the original graph, through which the structure and characteristics of the original graph can be better revealed. Assuming there is a planar graph G = (V, E), then the dual graph G of G * =(V * ,E * ), where V * is a collection of nodes: Each node Represents a road segment in the road network; E * is the set of edges: side Indicates that there is a connection relationship between the two corresponding road sections in the original graph. In the road network, the road network can be converted into a dual graph using the dual topology method, so as to better understand the structural and functional complexity of the road network. In the urban road network, each node of the dual graph corresponds to a stroke, and the edge of the dual graph corresponds to the intersection relationship between strokes. This conversion enables the structural features in the original graph to be better displayed in the dual graph, thereby better revealing the structure and function of the road network. As an implementation method of an embodiment of the present invention, in step S2, according to the road network dual model, the importance of urban roads is evaluated using length, connectivity, betweenness and proximity as measurement indicators, and the skeleton of the road network is extracted. The calculation method of node length, connectivity, proximity and betweenness is shown in Table 1.

[0031] Table 1

[0032]

[0033] The length, connectivity, proximity and betweenness of each road section are calculated, and the scores of the four indicators are weighted averaged by the entropy method to obtain a comprehensive score of the importance of each road section. The comprehensive score can reflect the importance and centrality of the road section in the entire road network; all road sections are sorted from high to low according to the comprehensive importance score, and the road with higher importance is selected as the selection result. As an implementation method of the embodiment of the present invention, in step S3, the fuzzy radiation center is extracted by the hot spot analysis method according to the skeleton line. Further, the road density reflects the density and traffic accessibility of the road network. The area with higher road density usually represents a transportation hub or a traffic agglomeration area, which is often a potential radiation center. At the same time, the POI density reflects the degree of concentration of commercial, cultural and social activities in the surrounding area. The area with higher POI density is usually a region with dense population and frequent activities, and may also be a potential radiation center. The comprehensive road density and POI density can more comprehensively reveal the characteristics of traffic and population activities in the region, which helps to identify potential radiation centers. In actual situations, the area with denser roads and more concentrated points of interest is often the core or hot spot of the city. To this end, the embodiment of the present invention establishes the following function model based on the road density and POI density:

[0034] D(m i )=ω Road D R (m i )+ω POI D P (m i )

[0035] Among them, D(m i ) is the overall density of the mesh, D R (m i ) represents mesh size m i The road density, D P (m i ) represents mesh size m i POI density, ω Road and ω POI is the weight.

[0036] Hot spot analysis is used to identify hot spots in spatial data, allowing users to discover clustering patterns in data sets. Through hot spot analysis, the city's fuzzy radiation center can be further identified from the comprehensive density score, providing an important reference for understanding urban transportation and activity centers. Local statistics It can be expressed as:

[0037]

[0038] Among them, x j is the attribute value of feature j, ω i,j is the spatial weight between features i and j, n is the total number of features, is x j The mean of x j The standard deviation of

[0039]

[0040] The Gi* statistic returned for each feature in the dataset is the z score. Using a 95% confidence interval, corresponding to z = 1.96, we select the mesh with a z score greater than 1.96 as the fuzzy radiation center. Figure 2 as shown in (a) and (b).

[0041] The fuzzy radiation center obtained after extraction may be more complex in graphics, and it needs to be appropriately simplified to highlight its main features. The simplification method is to reduce the dimension of the radiation center, that is, use the centroid of the extracted radiation center polygon to represent the radiation center point.

[0042] As an implementation of the embodiment of the present invention, in step S3, according to the skeleton line and the fuzzy radiation center, the radiation edge is extracted and linearized by a breadth-first search method.

[0043] Furthermore, in the standard circular radiation pattern, the radiation edge is a straight line segment radiating from the radiation center to the surroundings, representing the main traffic flow direction in the road network. Extracting and identifying the radiation edge can reveal the radiation structure and traffic flow pattern of the road network. The embodiment of the present invention selects the angle and the side length as the characteristic parameters of the radiation edge, and uses the breadth-first algorithm to search from the inside to the outside, thereby ensuring the accuracy of the search results; at the same time, by calculating the standard deviation of the angle and the side length as a discriminant model, the matching degree of the radiation edge relative to the standard circular radiation pattern can be better evaluated, thereby obtaining objective and accurate results.

[0044] The extraction process of radial edges includes: first, starting from the extracted fuzzy radial center, search outward along the topological structure of the road network to identify candidate radial edges. Specifically, a breadth-first search algorithm is used to identify all road segments directly connected to the fuzzy radial center, and these road segments are marked as candidate radial edges. After obtaining the candidate radial edges, they need to be checked for bifurcations. When there are multiple road segments that intersect or connect with each other in the candidate radial edges, the bifurcations must be processed. The usual processing method is to retain the higher-level main roads and delete the lower-level bifurcated roads to ensure that the extracted radial edges have higher clarity and continuity. According to the above steps, the range and shape of the radial edges can be determined, and finally a clear radial edge radiating outward can be obtained, such as Figure 3 As shown in (a), (b), (c) and (d).

[0045] The extraction results of the radiation edge are usually complex and have great differences from the standard circular radiation pattern, so the extracted pattern needs to be appropriately simplified to highlight the important characteristics of the pattern. Through the linearization operation, complex curves or polylines can be converted into straight line segments, thereby simplifying the graphic structure, reducing complexity, and making the graphics easier to display, analyze and process. To this end, the radiation edge is linearized on the basis that the radiation center has been converted into the centroid.

[0046] The centroid of the radiation center is the common endpoint of all the radiation edges, and all the radiation edges diverge from the centroid. In order to maintain the position and direction characteristics of the candidate radiation edges, the specific method of linearization is: for each radiation edge, determine a line segment, set the centroid of the radiation center as a fixed endpoint, and the other endpoint is the projection point of the original candidate radiation edge on the line segment, satisfying the minimum sum of the perpendicular distances from each inflection point on the original radiation edge to the line segment, such as Figure 4 shown.

[0047] Under ideal conditions, the lengths of the radiation sides of the standard annular radiation pattern are similar, and the angles between the radiation sides are equal, that is, the smaller the difference in the lengths of the radiation sides, the more uniform the distribution of the angles of the radiation sides, and the closer it is to the standard annular radiation pattern. A discrimination model is established based on the characteristics of the standard annular radiation pattern. By calculating the standard deviation of the length and angle distribution, the degree of approximation between the model and the standard annular radiation pattern can be intuitively reflected. Considering the angle distribution law and the difference in the length of the radiation sides, the discrimination model of the radiation sides is introduced:

[0048] S R (x,y)=ω Angle Anlge(x)+ω Length Length(y),S R ∈[0,1]

[0049] Among them, S R The closer the value is to 1, the closer the distribution of the radiating edge is to the standard annular radiation pattern; x is the standard deviation of the radiating edge angle; y is the standard deviation of the radiating edge length; Angle(x) is the discriminant function of the angle distribution law; Length(y) is the discriminant function of the degree of difference in the length of the radiating edge; ω Angle and ω Length is the weight.

[0050] Angle(x) is an indicator that uses the standard deviation of the angle to measure the uniformity of the angle distribution. When x is small, Angle(x) approaches 1; Length(y) is an indicator that uses the standard deviation of the length of the radiating side to measure the difference in the length of the target side. When y is small, Length(y) approaches 1. Since the units and sizes of the angle and length are different, normalization is required. According to the mathematical meaning of the standard deviation, when the angles of the radiating sides are equal, x = 0; when the lengths of the radiating sides are equal, y = 0. Based on this, a linear function is established:

[0051]

[0052] Among them, a is the maximum standard deviation of the angle under n sides; b is the maximum standard deviation of the side length under n sides; the values ​​of a and b change with the number of sides n.

[0053]

[0054] After the discriminant model is established, it is necessary to evaluate the combination of candidate radiating edges and calculate the score of the radiating edges belonging to the annular radiation pattern.

[0055] As an implementation manner of the embodiment of the present invention, a loop is extracted based on the skeleton line, the fuzzy radiation center and the radiation edge, and based on the angle and the turn of the connecting road segments.

[0056] Furthermore, in a ring-shaped radial pattern road network, the periphery of the radiation center usually contains several loops, which serve to connect the various radiating edges and are an important component of the road network. When identifying the ring-shaped radial pattern of a road network, the existence of these loop elements must be taken into account. The embodiment of the present invention extracts loops in the road network based on semantic information to ensure the accuracy of the extraction results to the greatest extent; compactness, deviation, area ratio, and perimeter ratio are selected as evaluation indicators for the loop, which can objectively and effectively evaluate the degree of proximity of the loop to the standard ring-shaped radial pattern. By identifying and analyzing the loops in the road network, it can help determine the radiation center and radiation edges of the road network, thereby revealing the overall structure and layout of the road network.

[0057] Loop extraction refers to identifying all possible loops from road network data. In real road networks, circular sections have good integrity, closure and coherence, that is, a loop is a closed path composed of a group of sections and has continuity in space. The embodiment of the present invention establishes the structural parameters of the circular pattern in a cognitive sense starting from the orientation (angle and turn) of the connecting sections, and sets constraints to extract the initial loop. The section angle refers to the angle between the intersection and the outer endpoints of adjacent sections, which is used to describe the smoothness of the connecting sections. Section turn refers to the deflection direction of the latter section compared to the previous section. Figure 5As shown in (a), according to the graphical features, the turning directions of the road segments in the loop should be consistent. From the geometric method, we know that if the line segments AB, BC, and CD are adjacent road segments in the loop and they are regarded as vectors, if they satisfy:

[0058]

[0059] This means that the direction of rotation from section AB to section BC is opposite to that from BC to CD, that is, the sections have inconsistent directions; otherwise, the directions of sections AB, BC, and CD are consistent. However, due to the accuracy error of the road data collection position and the actual distribution of the roads, there may be a situation where the direction of the section is inconsistent but the direction is consistent with the adjacent sections, such as Figure 5 (b) In order to overcome this defect, the embodiment of the present invention sets the angle threshold as another section control parameter. When the section angle is greater than this threshold, it can be considered that the adjacent sections are collinear, and the section steering consistency problem is no longer considered.

[0060] When the parameter threshold of the road section angle is set to 120°, the connection between the road sections can maintain good continuity. In summary, the loop extraction method adopted by the embodiment of the present invention includes the following steps: ① Establish an adjacency list of the road section and store the road sections connected to it; ② Search for adjacent road sections according to the angle priority principle, and add the arc segment IDs that meet the conditions to the first list; ③ Search for adjacent road sections according to the turn priority principle, and add the road section IDs that meet the conditions to the second list; ④ If the end point of the termination road section in the first list (or the second list) is the same as the starting point of the starting road section, all road section IDs of the loop are recorded in the first list (or the second list).

[0061] The loop in the annular radiation pattern has the following characteristics: ① It is circular or elliptical in shape; ② The position of the center of mass is close to the position of the center of mass of the radiation center; ③ Concentric rings are formed around the radiation center. The definition of loop characteristic parameters is considered from the above three aspects. At the same time, according to the principles of "pictures can be calculated" and "general acceptance", the embodiment of the present invention selects compactness, deviation, area ratio and perimeter ratio as characteristic parameters for identifying loops to ensure the accuracy and objectivity of the identification process. Table 2 shows the meaning and calculation method of these four parameters; Figure 6 As shown, these parameters are visualized. i is the loop centroid c μ The distance to the edge of the loop, D is d i A collection of is the average distance from the loop centroid to the loop boundary.

[0062] Table 2

[0063]

[0064] For each loop, the parameters defined in Table 2 are calculated, and the weights are constructed using the entropy method to calculate the score of the loop belonging to the circular radiation mode:

[0065] S C =ω1C+ω2D+ω3R a +ω4R p ,S C ∈[0,1]

[0066] Similarly, S C The closer the value of is to 1, the closer the distribution of the loop is to the standard circular radiation pattern. ω1, ω2, ω3, and ω4 are the weights of compactness, deviation, area ratio, and perimeter ratio, respectively.

[0067] As an implementation method of an embodiment of the present invention, in step S4, when evaluating the importance scores of the radial edges and loops, their comprehensive scores are calculated by a weight allocation method. The weight allocation method is to determine the importance of the radial edges and loops by an entropy method, and then perform a weighted summation of the radial edge scores and the loop scores according to the weight ratio to obtain a comprehensive score; the higher the comprehensive score, the closer the road network model is to the standard ring-shaped radial mode:

[0068] S=ω R S R +ω C S C ,S∈[0,1]

[0069] Among them, S is the comprehensive score; S R is the score of the radiating edge belonging to the annular radiation mode; S C is the score of the loop belonging to the annular radial mode; ω R ,ω C They are the weight of the importance of the radiating edge and the weight of the importance of the loop; the comprehensive score range is between 0 and 1, and the closer the score is to 1, the closer it is to the standard annular radiation pattern.

[0070] Based on the extraction of the road network skeleton line, the embodiment of the present invention successfully identifies the circular radiation pattern of the road network by calculating the radiation center, radiation edge and loop of the road network, analyzing several characteristic parameters of the radiation edge and loop and constructing weights using the entropy method. The innovation of the embodiment of the present invention is to use the skeleton line of the road network to effectively improve the efficiency and accuracy of identifying the road network pattern.

[0071] Embodiment 2:

[0072] The embodiment of the present invention further provides a ring-shaped radial road network pattern recognition device, comprising:

[0073] Establishing a module for establishing a road network dual model; wherein the road network dual model is used to visualize the topological relationship of the original road network;

[0074] The first extraction module is used to evaluate the importance of urban roads based on the road network dual model using length, connectivity, betweenness and proximity as measurement indicators, and extract the skeleton line of the road network;

[0075] The second extraction module is used to extract the fuzzy radiation center, radiation edge and loop of the road network according to the skeleton line; wherein the second extraction module extracts the fuzzy radiation center by a hot spot analysis method according to the skeleton line; the second extraction module extracts the radiation edge by a breadth-first search method according to the skeleton line and the fuzzy radiation center and performs linearization processing; the second extraction module extracts the loop according to the skeleton line, the fuzzy radiation center and the radiation edge based on the angle and turn of the connecting road section;

[0076] The third recognition module is used to evaluate the importance scores of the radiation edges and loops based on the weights to achieve ring radiation pattern discrimination.

[0077] Embodiment 3:

[0078] An embodiment of the present invention also provides a ring-shaped radial road network pattern recognition system, comprising: a memory and a processor, wherein the memory stores a computer program executed by the processor, and the computer program executes a ring-shaped radial road network pattern recognition method when executed by the processor.

[0079] Embodiment 4:

[0080] An embodiment of the present invention further provides a storage medium, on which a computer program is stored, and the computer program executes a method for identifying a ring-shaped radial road network pattern when running.

[0081] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.

Claims

1. A method for identifying a circular radial road network pattern, characterized in that: include: Step S1, establishing a road network dual model; wherein the road network dual model is used to visualize the topological relationship of the original road network; Step S2: According to the road network dual model, the importance of urban roads is evaluated using length, connectivity, betweenness and proximity as measurement indicators, and the skeleton line of the road network is extracted; Step S3, extracting the fuzzy radiation center, radiation edge and loop of the road network according to the skeleton line; wherein, according to the skeleton line, the fuzzy radiation center is extracted by a hot spot analysis method; according to the skeleton line and the fuzzy radiation center, the radiation edge is extracted by a breadth-first search method and linearized; according to the skeleton line, the fuzzy radiation center and the radiation edge, the loop is extracted based on the angle and turn of the connecting road section; Step S4: Evaluate the importance scores of the radiating edges and loops based on the weights to achieve ring-shaped radiation pattern discrimination.

2. A circular radial road network pattern recognition device, characterized in that: include: Establishing a module for establishing a road network dual model; wherein the road network dual model is used to visualize the topological relationship of the original road network; The first extraction module is used to evaluate the importance of urban roads based on the road network dual model using length, connectivity, betweenness and proximity as measurement indicators, and extract the skeleton line of the road network; The second extraction module is used to extract the fuzzy radiation center, radiation edge and loop of the road network according to the skeleton line; wherein the second extraction module extracts the fuzzy radiation center according to the skeleton line by a hot spot analysis method; the second extraction module extracts the radiation edge according to the skeleton line and the fuzzy radiation center by a breadth-first search method and performs linear processing; the second extraction module extracts the loop according to the skeleton line, the fuzzy radiation center and the radiation edge based on the angle and turn of the connecting road section; The identification module is used to evaluate the importance scores of the radiation edges and loops based on the weights to achieve ring radiation pattern discrimination.

3. A circular radial road network pattern recognition system, characterized in that: include: A memory and a processor, wherein the memory stores a computer program executed by the processor, and when the computer program is executed by the processor, the method for recognizing a ring-shaped radial road network pattern as claimed in claim 1 is executed.

4. A storage medium, characterized in that: The storage medium stores a computer program, which executes the ring-shaped radial road network pattern recognition method as claimed in claim 1 when running.

Images