Method for Evaluating Quality of Micro Map Road Network Based on Spatial Similarity

Abstract

The present invention discloses a method for evaluating the quality of a micro-map road network based on spatial similarity. First, the context information is combined with five spatial characteristics, namely shape, direction, distance, topological relationship, and one-dimensional line features, to calculate the spatial similarity between the micro-map data and the reference data. Then, the sample data is combined, and the objective weights are calculated using the entropy value method, and the average value is taken as the weight coefficient of each evaluation index. Finally, the extension cloud evaluation method is used to complete the comprehensive evaluation that combines the quantitative calculation and qualitative analysis of the micro-map data. The spatial similarity calculation method proposed by the present invention conforms to the characteristics of low precision and small content of the micro-map, and the evaluation method adopted helps to solve the problems of fuzziness, randomness, and incompatibility that occur in the evaluation process. The quality evaluation result is relatively accurate, providing a feasible method for the quality evaluation of the micro-map road network and facilitating the wide dissemination of the micro-map data.

Description

Technical Field

[0001] The present invention relates to the field of micro-map data quality evaluation, and more specifically to a method for evaluating the quality of a micro-map road network based on spatial similarity. Background Art

[0002] A micro-map is a new type of map for the general public generated in the self-media era, which can effectively make up for the deficiencies in the production and dissemination of traditional maps. Different from traditional maps, micro-maps have characteristics such as low cartographic accuracy, small data volume, and less expressed content. However, due to the lack of threshold requirements for mapmakers and the wide range of data sources, the quality of micro-maps is uneven, seriously affecting the popularity and credibility of micro-map applications.

[0003] Currently, common vector map data quality evaluation methods include the comparison and reference method, the matching evaluation method, and credibility measurement, etc. The comparison and reference method is to construct a reasonable evaluation model according to the selected evaluation indicators to achieve the quality evaluation of map data. The matching evaluation method is to establish the corresponding relationship between homonymous entities according to the matching results of map data, and analyze the differences between homonymous entities in combination with the selected evaluation indicators to achieve the quality evaluation of map data. However, for non-systematically deformed map data such as micro-maps, traditional map matching methods will have problems with low matching accuracy. The method based on credibility measurement is to reflect the degree of trustworthiness of data by mining relevant information of map data to complete the quality evaluation of map data. However, this method is only applicable to map data with a certain historical version. As newly emerged map data, micro-maps cannot meet this condition. To sum up, according to the characteristics of micro-map data, the comparison and reference method is a simple and effective map data quality evaluation method.

[0004] Spatial similarity plays an important role in spatial cognition. Currently, the spatial similarity calculation methods for evaluating the quality of line features are mostly for map data before and after comprehensive at different scales. The existing evaluation methods have a low overall evaluation level for micro-map data and are not applicable to the quality evaluation of micro-map road network data. Moreover, the influence of the shape similarity of line features on data quality evaluation is not considered in the existing methods. Summary of the Invention

[0005] In view of this, the present invention takes the vector data of the micro-map road network as the research object. According to the basic idea of shape context, five spatial characteristics of the shape, direction, distance, topological relationship, and one-dimensional line features of line features are selected to construct a spatial similarity measurement model considering context information, and on this basis, a method for evaluating the quality of the micro-map road network is proposed.

[0006] S1: Calculate the shape context of the endpoints of each road segment, count the probability of each point falling into each region, and calculate the shape similarity measurement value of the micro-map data and the reference data.

[0007] S2: Establish an eight-direction cone model, calculate the direction context of the endpoints of each road segment, count the probabilities of each point falling into each direction, and calculate the direction similarity metric value between the micro-map data and the reference data.

[0008] S3: Take any point as the coordinate origin, calculate the Manhattan distance from the remaining points to this point, and calculate the distance similarity metric value between the micro-map data and the reference data according to the distance context of each point.

[0009] S4: Define the topological relationship of the micro-map, calculate the topological context of each road segment, count the probabilities of each topological relationship, and calculate the topological similarity metric value between the micro-map data and the reference data.

[0010] S5: Compose a three-dimensional vector of the direction, length, and number of vertices of the road line to describe the one-dimensional line features. According to the line feature context of each road, calculate the line feature similarity metric value between the micro-map data and the reference data.

[0011] S6: Combine the sample data, calculate the objective weights of different combinations using the entropy method, and take the average as the weight coefficient of each metric value of spatial similarity.

[0012] S7: Use the calculated spatial similarity metric value as an evaluation index, and select the extension cloud evaluation method to achieve the comprehensive evaluation that combines the quantitative calculation and qualitative analysis of the micro-map data.

[0013] The spatial similarity calculation method considering context information proposed by the present invention conforms to the characteristics of low precision and small content of the micro-map. The adopted evaluation method can solve the problems of ambiguity, randomness, and incompatibility in the evaluation process. The quality evaluation result is relatively accurate, providing a feasible method for the quality evaluation of the micro-map road network, and is conducive to the wide dissemination of the micro-map data. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are only schematic diagrams of the present invention. For those of ordinary skill in the art, other drawings can be obtained according to the provided drawings without creative efforts.

[0015] Figure 1 It is a flowchart of the quality evaluation of the micro-map road network based on spatial similarity provided by the present invention.

[0016] Figure 2 It is a schematic diagram of shape context provided by the present invention.

[0017] Figure 3 It is a schematic diagram of direction context provided by the present invention.

[0018] Figure 4 Schematic diagram of distance context provided by the present invention.

[0019] Figure 5 Schematic diagram of topological relationship provided by the present invention.

[0020] Figure 6 Schematic diagram of the minimum area circumscribed rectangle provided by the present invention.

[0021] Figure 7 Reference map provided by the present invention.

[0022] Figure 8 Micro-map provided by the present invention.

[0023] Table 1 shows the evaluation results of micro-map quality.

[0024]

[0025] Detailed implementation manners

[0026] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0027] The embodiment of the present invention discloses a method for evaluating the quality of a micro-map road network based on spatial similarity, including:

[0028] Constructing a spatial similarity measurement model considering context information:

[0029] Step1: According to the shape context descriptor, calculate the shape context of the end-point data of each road segment in the micro-map to obtain an n×60 matrix.

[0030] Step2: Calculate the weighted average value of the shape context feature matrix by column to obtain the number of points falling into each region, and obtain the probability of falling into each region from the total number of points in each region.

[0031] Step3: Calculate the shape similarity of each region between the micro-map data and the reference data according to Equation (1).

[0032]

[0033] In the formula, Pa j , Pb jThey respectively represent the probabilities that the micro-map data and the reference data fall into each region obtained from the shape context matrix.

[0034] Step4: Calculate the weighted average of the similarities of each region to complete the shape similarity measurement between the micro-map data and the reference data.

[0035] Step5: Based on the idea of shape context, taking any road segment endpoint as the reference point, establish an eight-directional conical model, and count the number of the remaining n - 1 points in each direction region as the direction context of this point. Calculate the direction context of all points in turn to obtain an n×9 direction context feature matrix.

[0036] Step6: Calculate the weighted average of the direction context feature matrix by column, which represents the number of points falling into each direction. Calculate the probability of falling into each direction from the total number of points falling into each direction.

[0037] Step7: Obtain the similarities of each direction of the micro-map data and the reference data according to the probabilities of falling into each direction, as shown in Equation (2).

[0038]

[0039] In the formula, Pa j , Pb j They respectively represent the probabilities that the points obtained from the direction context feature matrix of the micro-map data and the reference data fall into each direction.

[0040] Step8: Calculate the weighted average of the similarities of each direction to complete the direction similarity measurement between the micro-map data and the reference data.

[0041] Step9: Taking any road segment endpoint as the coordinate origin, calculate the Manhattan distance from the remaining points to this point to describe the separation degree between two point features, and obtain the distance context of the points. Calculate the distance context of each point in the point set in turn to obtain an n×(n - 1) distance context feature matrix.

[0042] Step10: Calculate the weighted average of the distance context of each point, which is expressed as the distance context measurement value of each point. Calculate the weighted average of the distance context measurement values of each point to represent the distance context feature measurement value of this data, and complete the distance similarity measurement between the micro-map data and the reference data using Equation (3).

[0043]

[0044] In the formula, D a and D b They respectively represent the distance context feature measurement values of the micro-map data and the reference data.

[0045] Step11: Take any road segment as the reference object. According to the defined topological relationship, judge the topological relationship between the remaining line segments and this segment and make statistics. Calculate the topological context of all line segments in turn to obtain an n×4 matrix, which describes the topological context characteristics of the data.

[0046] Step12: Calculate the weighted average value of the topological context matrix by column, which is used to represent the total number of each topological relationship. Obtain the probability of each topological relationship from the total number of each statistical topological relationship. Calculate the similarity of the four topological relationships between the micro-map data and the reference data according to the probability of each topological relationship, as shown in Equation (4).

[0047]

[0048] In the formula, Pa j and Pb j respectively represent the probabilities of each topological relationship obtained from the topological context feature matrix of the micro-map data and the reference data.

[0049] Step13: Calculate the weighted average value of the similarities of the four topological relationships to complete the calculation of the topological similarity between the micro-map data and the reference data.

[0050] Step14: Calculate the minimum area circumscribed rectangle of each road line. Take the rotation angle and the major axis of the minimum area circumscribed rectangle as the direction θ and length x of the line feature. Combine the number of vertices n of the line feature to describe the one-dimensional line feature of the road network, and represent it with a three-dimensional vector (θ, x, n).

[0051] Step15: Select any line segment as the reference object. Calculate the Manhattan distance from the feature vector of the remaining line segments to the feature vector of this segment, describe the difference in the characteristics of the two road lines, and use it as the feature context of this segment. Calculate the weighted average value of the feature context of each line segment as the measurement value of the feature context of each line segment.

[0052] Step16: Use the weighted average value of the feature context measurement values of each line segment to represent the measurement value of the one-dimensional line feature context of this data, and complete the measurement of the line feature similarity between the micro-map data and the reference data using Equation (20).

[0053]

[0054] In the formula, L a and L b respectively represent the measurement values of the one-dimensional line feature context of the micro-map data and the reference data.

[0055] Step17: Combine the sample data: Suppose there are n sample data, and m sample data are selected from them, there are kinds of combination results

[0056] Step18: Calculate the weight coefficients of each combination of sample data using the entropy method, and determine the weight coefficients of each evaluation index by taking the average value.

[0057] Micro-map road network data quality evaluation:

[0058] Step19: Take the calculated spatial similarity metric value as an evaluation index, establish a micro-map data quality evaluation grade, and use the extension cloud model to calculate the correlation degree between each evaluation index of the micro-map data and different evaluation grades to complete the quality evaluation of the micro-map road network data.

[0059] From Figure 7 The data for verification shows that the quality evaluation results are relatively accurate. The spatial similarity calculation method proposed in this paper conforms to the characteristics of low precision and less content of micro-maps, can better divide appropriate grades for micro-map data, and provides a feasible method for the quality evaluation of micro-map road networks.

[0060] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present invention. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined by this solution can be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention will not be limited to these embodiments shown by this solution, but rather should conform to the widest scope consistent with the principles and novel features disclosed by the present invention.

Claims

1. A method for evaluating the quality of a micro-map road network based on spatial similarity, which includes two parts: constructing a spatial similarity measurement model considering context information and evaluating the quality of micro-map road network data: The steps for constructing a spatial similarity measurement model considering context information are as follows: S1: Calculate the shape context of the endpoints of each road segment, count the probability of each point falling into each region, and calculate the shape similarity measurement value between the micro-map data and the reference data; S2: Establish an eight-direction cone model, calculate the direction context of the endpoints of each road segment, count the probability of each point falling into each direction, and calculate the direction similarity measurement value between the micro-map data and the reference data; S3: Take any point as the coordinate origin, calculate the Manhattan distance from the remaining points to this point, and calculate the distance similarity measurement value between the micro-map data and the reference data according to the distance context of each point; S4: Define the topological relationship of the micro-map, calculate the topological context of each road segment, count the probability of each topological relationship, and calculate the topological similarity measurement value between the micro-map data and the reference data; S5: Use a three-dimensional vector composed of the direction, length, and number of vertices of the road line to describe the one-dimensional line feature, and calculate the line feature similarity measurement value between the micro-map data and the reference data according to the line feature context of each road; S6: Combine the sample data, use the entropy method to calculate the objective weights of different combinations, and take the average value as the weight coefficient of each spatial similarity measurement value; The steps for evaluating the quality of micro-map road network data are as follows: S7: Use the calculated spatial similarity measurement value as an evaluation index, and adopt the extension cloud evaluation method to complete the comprehensive evaluation that combines the quantitative calculation and qualitative analysis of the micro-map data; S8: End.

2. The method for evaluating the quality of a micro-map road network based on spatial similarity according to claim 1, wherein, in steps S1 to S4, according to the basic idea of shape context, select five spatial characteristics of the shape, direction, distance, topological relationship, and one-dimensional line feature of the line element, and calculate the spatial similarity measurement value considering context information.

3. The method for evaluating the quality of a micro-map road network based on spatial similarity according to claim 1, wherein, calculate the objective weight of the sample data combination result in step S5.

4. The method for evaluating the quality of a micro-map road network based on spatial similarity according to claim 1 or claim 2, wherein, adopt the extension cloud evaluation method to complete the comprehensive quality evaluation of the micro-map road network data in step S7.

Images