A method for matching road intersections based on the topological relationships of adjacent objects

Abstract

This invention discloses a road intersection matching method that combines the topological relationships of adjacent features, belonging to the field of remote sensing image processing technology. It includes the following steps: extracting a set of road intersections from the semantic information of the entire city's features; calculating three topological vectors for each road intersection; finding road intersections in the road intersection set with the same number of adjacent categories as the road intersection to be matched, forming a candidate set of road intersections; calculating the three topological vectors of the road intersection to be matched; calculating the topological vector distance between the road intersection to be matched and each road intersection in the candidate set; finding the road intersection that minimizes the three topological vector distances; if two road intersections have the same number, the match is successful. This invention can fully utilize the semantic features and spatial distribution relationships of features surrounding road intersections to construct a matching model for road intersections, providing support for the fusion application of heterogeneous geographic data.

Description

Technical Field

[0001] This invention belongs to the field of geographic information processing technology, specifically relating to a road intersection matching method that combines the topological relationships of adjacent objects. Background Technology

[0002] With the rapid development of geospatial information technology, the methods for acquiring the location and morphology of ground features have become increasingly diversified, and information obtained through different methods has different advantages and errors. Therefore, in order to comprehensively utilize information from different data sources for the same target, it is necessary to register the various data sources. As an important transportation facility within cities, the location and spatial distribution of road intersections are of great significance for urban planning and the matching of multi-source data, and are often used as key ground feature elements for matching heterogeneous data.

[0003] However, with the development of urban planning and infrastructure, urban planning and land resource management departments have built databases of semantic and locational attribute information of urban features, resulting in relatively complete urban feature distribution maps. Meanwhile, the development of edge computing has made it very easy to obtain semantic information of different features at the edge, but the geographic information and feature attributes it possesses are usually limited, and the location errors can be significant, posing many difficulties for the fusion of heterogeneous data. Road intersections have high recognizability and obvious grayscale features in urban remote sensing images. Most existing research and patents rely on image features for identification and registration, while some studies perform direct feature-level matching based on low-level image grayscale features. Therefore, based on urban feature semantic databases, how to utilize the semantic information and spatial topological features of features around road intersections to achieve rapid heterogeneous geographic data matching has become an important research direction. Summary of the Invention

[0004] The purpose of this invention is to provide a road intersection matching method that combines the topological relationship of adjacent objects. This method can make full use of the semantic features and spatial distribution relationship of the ground objects around the road intersection to construct a matching model of the road intersection, thus providing support for the fusion application of heterogeneous geographic data.

[0005] The technical solution adopted in this invention is as follows:

[0006] A road intersection matching method that incorporates the topological relationships of adjacent objects includes the following steps:

[0007] Step 1: Obtain semantic information of geographical features within the entire city and extract the set I of road intersections. m Let the i-th road intersection be ru i ;

[0008] Step 2, obtain the road intersection rui The semantic and locational attributes of adjacent land features of various types, including the semantic attributes of road intersections. i Total number of surrounding land features k i and road intersections i The total number t of adjacent land features of type j j Location attributes include road intersections. i Adjacent to the g-th type of land feature in class s s Relative coordinates of individual ground features And there are:

[0009]

[0010] In the formula, For road intersections i The center coordinates, For the gth s The center coordinates of each ground feature;

[0011] Step 3, Convert to polar coordinates in, Represents the g-th type of land cover in the s-th category. s The distance between the center point of each feature and the center point of the intersection, i.e., the polar radius. Indicates the polar angle;

[0012] Step 4: Calculate the intersection ru based on the spatial distribution of various land features. i The three topological vectors: variance topological vector Mean topological vector Median topological vector

[0013] Step 5: Obtain the road intersection re to be matched, and obtain the semantic and location attributes of each type of land feature adjacent to re. The semantic attributes include the total number n of adjacent land features around the road intersection re, and the total number h of land features of type a adjacent to road intersection re. a Location attributes include the road intersection re adjacent to the hth feature of type a. a Relative coordinates of individual ground features

[0014] Step 6, set up I at the road intersection m Find road intersections with the same number of adjacent categories as the road intersection to be matched, and form a candidate set I of road intersections. m1 ;

[0015] Step 7, Convert to polar coordinates Calculate the variance topology vector of the road intersection re to be matched Mean topological vector Median topological vector

[0016] Step 8: Calculate the intersection re to be matched and the candidate set I of road intersections. m1 At each road intersection e Topological vector distance DV(re,rs) e ), DM(re,rs e ) and DO(re,rs e );

[0017] Step 9, find the values ​​that make DV(re,rs) equal to... e ), DM(re,rs e ) and DO(re,rs e Take the minimum value of I m1 If there are two road intersections re1, re2, and re3 with the same number, then the road intersection with the same number is the successfully matched road intersection re_rs.

[0018] Furthermore, in step 3, polar coordinates The calculation method is as follows:

[0019]

[0020]

[0021] Here, actan() represents the arctangent function.

[0022] Furthermore, in step 4, the road intersection ru i variance topological vector Mean topological vector Median topological vector They are respectively:

[0023]

[0024]

[0025]

[0026] Here, Var(), Mean(), and Mode() respectively calculate the variance, mean, and median of the array, k i It is a road intersection. i The total number of surrounding land features; and They represent road intersections rui The set of polar radius and polar angle of each feature entity in the surrounding j-th type of land cover, i.e.:

[0027]

[0028]

[0029] In the formula, t j Indicates road intersection ru i The total number of adjacent land features of type j.

[0030] Furthermore, in step 7, the variance topological vector of the road intersection re Mean topological vector Median topological vector They are respectively:

[0031]

[0032]

[0033]

[0034] Here, Var(), Mean(), and Mode() respectively calculate the variance, mean, and median of the array, k i It represents the total number of land feature types around the road intersection re; and Let represent the set of polar radius and the set of polar angle of each feature entity of type a surrounding the road intersection re, respectively.

[0035]

[0036]

[0037] In the formula, h a This represents the total number of land features of type a adjacent to the road intersection re.

[0038] Furthermore, in step 8, the topological vector distance DV(re,rs) e ), DM(re,rs e ) and DO(re,rs e The calculation method for ) is as follows:

[0039]

[0040]

[0041]

[0042] Where q represents the serial number of the adjacent objects around the road intersection.

[0043] The present invention has the following beneficial effects:

[0044] (1) This invention proposes a road intersection matching method that combines the topological relationship of adjacent objects, effectively utilizing the semantic information of adjacent objects at road intersections to match the most similar road intersections.

[0045] (2) The method of the present invention can effectively match road intersections in the city, providing effective technical support for the matching and fusion of multi-source heterogeneous data. Attached Figure Description

[0046] Figure 1 This is a flowchart of a road intersection matching method that incorporates the topological relationships of adjacent objects. Detailed Implementation

[0047] The present invention will now be described in further detail with reference to the accompanying drawings and specific embodiments.

[0048] A road intersection matching method that incorporates the topological relationships of adjacent objects includes the following steps:

[0049] Step 1: Obtain semantic information of geographical features within the entire city and extract the set I of road intersections. m Let the i-th road intersection be ru i Road intersection set I m It can be represented as:

[0050] I m ={ru1,ru2,...,ru m}

[0051] In the formula, m represents the set of road intersections I. m The number of road intersections in the middle.

[0052] Step 2, obtain the road intersection ru i The semantic and locational attributes of adjacent land features of various types, including the semantic attributes of road intersections. i Total number of surrounding land features k i and road intersections i The total number t of adjacent land features of type j j Location attributes include road intersections. i Adjacent to the g-th type of land feature in class s s Relative coordinates of individual ground features And there are:

[0053]

[0054] In the formula, For road intersections i The center coordinates, For the gth s The center coordinates of each ground feature.

[0055] Step 3, Convert to polar coordinates

[0056]

[0057]

[0058] Here, actan() represents the arctangent function.

[0059] Step 4: Calculate the intersection ru based on the spatial distribution of various land features. i The three topological vectors and

[0060]

[0061]

[0062]

[0063] Here, Var(), Mean(), and Mode() respectively calculate the variance, mean, and median of the array, and k1 is the intersection of roads. i The total number of types of land features in the surrounding area.

[0064] and They represent road intersections ru i The set of polar coordinates of the j-th type of surrounding land features is calculated as follows:

[0065]

[0066]

[0067] In the above formula, t j Indicates road intersection ru i The total number of adjacent land features of type j.

[0068] Step 5: Obtain the road intersection re to be matched, and obtain the semantic and location attributes of each type of adjacent land feature. The semantic attributes include the total number n of adjacent land features around the road intersection re, and the total number h of type a land features adjacent to road intersection re. aLocation attributes include the road intersection re adjacent to the hth feature of type a. a Relative coordinates of individual ground features

[0069] Step 6, gather at the intersection of urban roads, I m Find the candidate set of road intersections I that matches the number of re-adjacent categories of the road intersection to be matched. m1 ;

[0070] Step 7, Convert to polar coordinates Calculate the topology vector of the road intersection re to be matched and

[0071]

[0072]

[0073]

[0074] Here, Var(), Mean(), and Mode() respectively calculate the variance, mean, and median of the array, k i It represents the total number of land feature types around the road intersection re; and Let represent the set of polar radius and the set of polar angle of each feature entity of type a surrounding the road intersection re, respectively.

[0075]

[0076]

[0077] In the formula, h a This represents the total number of features of type a adjacent to the road intersection re;

[0078] Step 8: Calculate the candidate set of road intersections I and the road intersections to be matched. m1 At each road intersection e Topological vector distance DV(re,rs) e ), DM(re,rs e ) and DO(re,rs e ):

[0079]

[0080]

[0081]

[0082] Where q represents the serial number of the features around the road intersection.

[0083] Step 9, find the values ​​that make DV(re,rs) equal to... e ), DM(re,rs e ) and DO(re,rs e Take the minimum value of I m1 The set of road intersections in the given information is re1, re2, re3:

[0084]

[0085]

[0086]

[0087] Where V, M, and O represent the distances based on the topological vector distance DV(re,rs), respectively. e ), DM(re,rs e ) and DO(re,rs e The intersection number that is most similar to the intersection re is calculated.

[0088] If there are two or more candidate road intersections with the same number among re1, re2, and re3, it means that the road intersection matching is successful, and the final matched road intersection re_rs is obtained.

[0089] The road intersection matching method proposed in this invention can effectively utilize the attributes and spatial topological relationships of adjacent objects at road intersections to construct semantic information of adjacent object categories and spatial topological vector matching rules, which can provide effective technical support for the matching and fusion of multi-source geographic data.

Claims

1. A road intersection matching method combining the topological relationships of adjacent objects, characterized in that, Includes the following steps: Step 1: Obtain semantic information of geographical features within the entire city and extract the set I of road intersections. m Let the i-th road intersection be ru i ; Step 2, obtain the road intersection ru i The semantic and locational attributes of adjacent land features of various types, including the semantic attributes of road intersections. i Total number of adjacent object types k i and road intersections i The total number t of land features adjacent to the j-th type of land feature j Location attributes include road intersections. i Adjacent to the g-th type of land feature in class s s Relative coordinates of individual ground features And there are: In the formula, For road intersections i The center coordinates, For the gth type of land cover in the s-th category s The center coordinates of each ground feature; Step 3, Convert to polar coordinates ,in, Represents the g-th type of land cover in the s-th category. s The center point of each geographical entity and the road intersection ru i The distance from the center point, i.e., the polar radius. Indicates the polar angle; Step 4: Calculate the intersection ru based on the spatial distribution of various land features. i The three topological vectors: variance topological vector Mean topological vector Median topological vector : in, , and These represent the variance, mean, and median of the array, respectively, where k i It is a road intersection. i The total number of adjacent object types; and They represent road intersections ru i The set of polar radius and polar angle of polar coordinates of all features adjacent to feature type j, i.e.: In the formula, t j Indicates road intersection ru i The total number of land features adjacent to type j; Step 5: Obtain the road intersection re to be matched, and obtain the semantic and location attributes of each type of land feature adjacent to re. The semantic attributes include the total number n of land feature types adjacent to the road intersection re, and the total number h of land feature entities of type a adjacent to the road intersection re. a Location attributes include the road intersection re adjacent to the hth feature of type a. a Relative coordinates of individual ground features ; Step 6, set up I at the road intersection m Find road intersections with the same number of adjacent object types as the road intersection to be matched (re), and form a candidate set of road intersections I. m1 ; Step 7, Convert to polar coordinates Calculate the variance topology vector of the road intersection re to be matched. Mean topological vector Median topological vector ; Step 8: Calculate the intersection re to be matched and the candidate set I of road intersections. m1 Every road intersection Topological vector distance , and : Where q represents the serial number of the type of adjacent object at the road intersection; Step 9, find the ones that make , and Take the minimum value of I m1 Given a set of road intersections re1, re2, and re3, if two road intersections in re1, re2, and re3 have the same number, then the road intersection with the same number is the successfully matched road intersection re_rs.

2. The road intersection matching method based on the topological relationship of adjacent objects according to claim 1, characterized in that, In step 3, polar coordinates The calculation method is as follows: Here, actan() represents the arctangent function.

3. The road intersection matching method based on the topological relationship of adjacent objects according to claim 2, characterized in that, The variance topological vector of the road intersection re in step 7 Mean topological vector Median topological vector They are respectively: in, , and These represent the variance, mean, and median of the array, respectively, where n is the total number of adjacent object types at the road intersection re. and Let represent the set of polar radius and the set of polar angle of each feature entity in class a adjacent to the road intersection re, respectively. In the formula, h a This represents the total number of land features of type a adjacent to the road intersection re.

Images