Map navigation data efficient processing system based on cloud storage

Abstract

The invention relates to the technical field of data processing, in particular to a map navigation data efficient processing system based on cloud storage, the system comprises a processor and a memory, and the processor executes a computer program stored in the memory to realize the following steps: analyzing the distribution condition of position data points on a navigation map; and acquiring a low-density set and a high-density set based on an analysis result, compressing data points in the high-density set by using a geometric transformation coding compression algorithm to obtain compressed data, and compressing data points in the low-density set by using an LZ compression algorithm to obtain compressed data. And the data compression effect can be improved, and the data transmission efficiency can also be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing, and in particular to an efficient processing system for map navigation data based on cloud storage. Background Art

[0002] With the rapid development of smart phones, car navigation devices, unmanned driving technology, etc., the application of navigation maps is becoming more and more extensive. At present, road conditions are usually checked and driving routes are planned based on navigation maps. At present, the location data points of objects on the road are usually transmitted to the data analysis and processing center or module of the navigation map in real time, and then the road conditions are updated in real time or the driving routes are planned in real time through the data processing analysis results of the data analysis and processing module. In addition, in general, the amount of data to be transmitted is relatively large, and in order to improve the transmission efficiency and enable the navigation map to plan a better driving or walking route or to update the road conditions in a timely manner, a data compression algorithm is usually used to compress the location data points to be transmitted, and then the compressed data is transmitted.

[0003] However, when compressing data, a compression method is usually used for compression, that is, any one of the geometric transform coding compression algorithm and the LZ compression algorithm is usually selected. However, since the compactness between data points is different, and different algorithms have different compression effects on data of different densities, if any compression algorithm is directly randomly selected to compress the data, this compression method will lead to poor compression effect. For example, in the data point set that needs to be compressed, if the compactness between some data points is high, then if the LZ compression algorithm is selected for compression at this time, the compression effect of this part of the data points will be relatively poor. If the compactness between some data points is low, then if the geometric transform coding compression algorithm is selected for compression at this time, the compression effect of this part of the data points will also be relatively poor. The main reason is that the geometric transform coding compression algorithm can achieve a higher compression effect in high-density areas by making full use of the redundancy and structural characteristics in spatial data, and is especially suitable for spatial data with regularity and compact distribution. The LZ compression algorithm can effectively compress data in low-density areas through dictionary replacement technology, identify repeated byte sequences to reduce redundancy, and thus improve storage efficiency. Therefore, how to improve the compression effect has become an urgent problem to be solved. Summary of the invention

[0004] In order to solve the above problems, the present invention provides a system for efficiently processing map navigation data based on cloud storage, and the technical solutions adopted are as follows: An embodiment of the present invention provides a system for efficiently processing map navigation data based on cloud storage, comprising a processor and a memory, wherein the processor executes a computer program stored in the memory to implement the following steps: Get location data points on the navigation map; The location data points on the navigation map are all recorded as first data points, and all the first data points are clustered using a clustering algorithm whose clustering parameters are initial clustering parameters, and the cluster clusters obtained after the clustering are completed are all recorded as first cluster clusters; according to the cluster center of each first cluster cluster, the core set corresponding to each first cluster cluster is obtained, and each is recorded as the first core set; according to all the data points in each first core set and the initial clustering parameter, the compactness representation value of each first core set is obtained; Determine whether the compactness representation value of each of the first core sets is greater than the first compactness threshold, if so, record the corresponding first core set as a high-density set, otherwise, record the corresponding first core set as a low-density set; Obtain the intersection area between any two first clusters, and record them as the first intersection area, determine whether the total number of first data points in all the first intersection areas is not greater than a preset number threshold, if so, record the set composed of all first data points that do not belong to the first core set as a low-density set, otherwise, record all first data points that do not belong to the first core set as second data points, and adjust the initial clustering parameters, continue to cluster the second data points according to the adjusted clustering parameters, and continue to obtain low-density sets and high-density sets according to the clustering results; The data points in the high-density set are compressed using a geometric transformation coding compression algorithm to obtain compressed data, and the data points in the low-density set are compressed using an LZ compression algorithm to obtain compressed data.

[0005] Beneficial effect: The present invention first obtains the location data points on the navigation map; then the location data points on the navigation map are all recorded as the first data points, and all the first data points are clustered using a clustering algorithm whose clustering parameters are the initial clustering parameters. The clustering clusters obtained after the clustering is completed are all recorded as the first clustering clusters, and according to the clustering center of each first clustering cluster, the core set corresponding to each first clustering cluster is obtained and recorded as the first core set, and according to all the data points in each first core set and the initial clustering parameters, the compactness characterization value of each first core set is obtained; then, it is determined whether the compactness characterization value of each first core set is greater than the first compactness threshold value, and if so, the corresponding first core set is recorded as a high-density set, otherwise, the corresponding first core set is recorded as a low-density set. ; Then obtain the intersection area between any two first clusters, and record them as the first intersection area, determine whether the total number of first data points in all first intersection areas is not greater than the preset number threshold, if so, then record the set composed of all first data points that do not belong to the first core set as a low-density set, otherwise, then record all first data points that do not belong to the first core set as second data points, and adjust the initial clustering parameters, continue to cluster the second data points according to the adjusted clustering parameters, and continue to obtain low-density sets and high-density sets according to the clustering results; finally, use the geometric transform coding compression algorithm to compress the data points in the high-density set to obtain compressed data, and use the LZ compression algorithm to compress the data points in the low-density set to obtain compressed data. Furthermore, the present invention can improve the data compression effect by dividing the data into a low-density set and a high-density set, and then compress the low-density set and the high-density set using different compression algorithms respectively, thereby also improving the transmission efficiency. That is, the transmission efficiency of the compressed data when it is subsequently transmitted to the data analysis and processing module in the navigation map system can be improved, and the efficiency and rationality of the route planning of the navigation map system can also be improved, and more timely road condition updates can be guaranteed. BRIEF DESCRIPTION OF THE DRAWINGS

[0006] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0007] Figure 1 The present invention is a flowchart of a method for efficiently processing map navigation data based on cloud storage. DETAILED DESCRIPTION

[0008] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field belong to the scope of protection of the embodiments of the present invention.

[0009] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0010] This embodiment provides a system for efficiently processing map navigation data based on cloud storage, including a processor and a memory, wherein the processor executes a computer program stored in the memory to implement a method for efficiently processing map navigation data based on cloud storage, such as Figure 1 As shown, the efficient processing method of map navigation data based on cloud storage includes the following steps: Step S001, obtaining a location data point on a navigation map.

[0011] The main purpose of this embodiment is to improve the compression effect of location data points to improve data transmission efficiency, so that the navigation map can plan a more reasonable driving or walking route or can update the road conditions in a timely manner; and for the convenience of analysis, this embodiment takes the compression process of location data points of objects on all roads in any area as an example for analysis, such as the area can refer to any city, and the area is recorded as a monitoring area.

[0012] First, all roads in the monitoring area are recorded as monitoring roads. Then, at the current monitoring time, the location data of all monitoring objects located on the monitoring roads are obtained. The location data of the monitoring objects mainly include longitude and latitude, and the monitoring objects include but are not limited to vehicles and people. However, the monitoring objects in this embodiment are required to be objects whose location data can be received by the map navigation, that is, the monitoring objects carry networked location acquisition devices or the location information of the monitoring objects can be obtained through remote networked monitoring devices. If a vehicle on a monitoring road can obtain location information through GPS devices and vehicle-mounted sensors, then this vehicle is a monitoring object. If an animal located on a monitoring road not only does not carry a networked location acquisition device, but all remote networked monitoring devices cannot locate its location information, then the animal is not a monitoring object.

[0013] And because the monitored objects carry networked location acquisition devices or remote networked monitoring devices that can obtain the location information of the monitored objects, the navigation map can receive the location data of these monitored objects in real time, and map the received location data one by one to the navigation map, so that at the current monitoring time, the location data points of all monitored objects can be obtained on the navigation map. In this embodiment, the dimension of the location data point is two-dimensional, so the horizontal coordinate value of the location data point is longitude and the vertical coordinate value is dimension; as other real-time methods, the location data point can also be three-dimensional, consisting of longitude, latitude and altitude respectively.

[0014] Therefore, through the above process, all location data points on the navigation map at the current moment can be obtained, and usually the number of location data points on the navigation map at the current monitoring moment is relatively large; in addition, these location data points will also be stored based on cloud storage; it should be noted that the location data points that appear subsequently are all location data points at the current monitoring moment, and the map navigation can update the road conditions in real time by analyzing these location data points, and plan reasonable routes for users using the map navigation.

[0015] Step S002, compressing the location data points on the navigation map to obtain compressed data.

[0016] Moreover, due to different times or different roads being located in different areas, the positions or quantities of monitored objects on different roads at different times are different, and different compression algorithms have different compression effects on areas with different compactness. Therefore, in order to improve the compression effect, this embodiment analyzes the compactness of data points in different areas, and adaptively selects a compression algorithm based on the analysis results, thereby improving the compression effect. The specific compression process is as follows: In this embodiment, all location data points on the navigation map are first recorded as first data points, and then initial clustering parameters are obtained, and all first data points are clustered using a clustering algorithm with clustering parameters as initial clustering parameters, and the clustering clusters obtained after the clustering is completed are recorded as first clustering clusters; then, based on the clustering center of each first clustering cluster, the core set corresponding to each first clustering cluster is obtained, and the core sets corresponding to all first clustering clusters are recorded as first core sets; then, based on all first data points in each first core set and the initial clustering parameters, the compactness characterization value of each first core set is obtained.

[0017] After obtaining the compactness characterization value of each first core set, the first compactness threshold is obtained, and then it is determined whether the compactness characterization value of each first core set is greater than the first compactness threshold. If so, it is determined that the distribution of data points in the corresponding first core set is high-density, and all first core sets whose compactness characterization value is greater than the first compactness threshold are recorded as high-density sets. Otherwise, it is determined that the distribution of data points in the corresponding first core set is low-density, and all first core sets whose compactness characterization value is not greater than the first compactness threshold are recorded as low-density sets.

[0018] And based on the above analysis, it can be known that the above process only determines whether the data points in the first core set belong to high-density distribution or low-density distribution, and does not analyze the data points other than the first core set. Therefore, this embodiment next obtains the intersection area between any two first clusters, and records them as the first intersection area, and determines whether the total number of first data points in all first intersection areas is not greater than the preset number threshold. If so, it is determined that the distribution between the data points outside the first core set is relatively discrete, then the set composed of all first data points that do not belong to the first core set is recorded as a low-density set. Otherwise, it is determined that it is also necessary to analyze the density of the data points other than the first core set, that is, if the total number of first data points in all first intersection areas is not less than the preset number threshold, then all first data points that do not belong to the first core set are recorded as second data points, and the initial clustering parameters are adjusted, and all second data points continue to be clustered according to the adjusted clustering parameters, and low-density sets and high-density sets continue to be obtained according to the clustering results.

[0019] The initial clustering parameters are adjusted, all the second data points are clustered according to the adjusted clustering parameters, and the process of obtaining low-density sets and high-density sets according to the clustering results is as follows: First, the set consisting of all the second data points in the navigation map is recorded as the second set, and the compactness representation value of the second set is obtained. Then, according to the compactness representation value of the second set and the first compactness threshold, the initial clustering parameter is adjusted to obtain the first adjusted clustering parameter; then, all the second data points are clustered using the clustering algorithm whose clustering parameter is the first adjusted clustering parameter, and the clustering clusters obtained after the clustering is completed are recorded as second cluster clusters; then, according to the clustering center of each second cluster cluster, the core set corresponding to each second cluster cluster is obtained, and each is recorded as the second core set; then, according to all the data points in each second core set and the first adjusted clustering parameter, the compactness representation value of each second core set is obtained; then, it is determined whether the compactness representation value of each second core set is greater than the second compactness threshold. If so, The second core set should be recorded as a high-density set, otherwise, the corresponding second core set will be recorded as a low-density set; then the intersection area between any two second clusters is obtained, and they are all recorded as second intersection areas; then it is determined whether the total number of second data points in all second intersection areas is not greater than the preset number threshold, if so, the set composed of all second data points that do not belong to the second core set is recorded as a low-density set, otherwise, all second data points that do not belong to the second core set are recorded as third data points, and the first adjustment clustering parameter is adjusted, and all third data points are continued to be clustered according to the adjusted clustering parameter, and low-density sets and high-density sets are obtained again according to the clustering results, and so on, until all the obtained low-density sets and high-density sets contain all the location data points on the navigation map and stop.

[0020] In this embodiment, the clustering algorithm used when clustering data points is the DBSCAN clustering algorithm, and the initial clustering parameters in this embodiment include the initial radius R and the initial minimum number of points M. In specific applications, the implementer can set the values ​​of R and M according to actual conditions, or set R and M to empirical values. For example, in this embodiment, R can be set to 2 and M can be set to 25. It should be noted that the initial minimum number of points refers to the minimum number of points in the neighborhood of R when clustering, that is, when clustering, if a point contains at least M points in the neighborhood of radius R, this point is called a core point. The area where the core point is located is a data-intensive area and can constitute the core of the cluster. In addition, in this embodiment, the process of clustering data points using the DBSCAN clustering algorithm when the clustering parameters are known is a well-known technology, so this embodiment will not be described in detail.

[0021] In this embodiment, the process of obtaining the cluster center of a cluster is as follows: for any cluster, the horizontal coordinate of the cluster center of the cluster is the mean of the horizontal coordinate values ​​of all data points in the cluster, and the vertical coordinate of the cluster center of the cluster is the mean of the vertical coordinate values ​​of all data points in the cluster.

[0022] In this embodiment, according to the cluster center of each first cluster, the specific process of obtaining the core set corresponding to each first cluster is as follows: for any first cluster A1: among all the first data points except the first cluster A1, obtain a first data point closest to the first cluster A1, and record it as the out-cluster nearest neighbor of the first cluster A1; then obtain the Euclidean distance between the cluster center of the first cluster A1 and the out-cluster nearest neighbor of the first cluster A1, and record it as the out-cluster nearest neighbor distance of the first cluster A1. Distance; then obtain a circle with the cluster center of the first cluster A1 as the center and the outer neighbor distance of the first cluster A1 as the radius, and record the area contained in the circle as the core area corresponding to the first cluster A1; then obtain all the first data points belonging to the first cluster A1 and located in the core area corresponding to the first cluster A1, and record the set formed by all the first data points belonging to the first cluster A1 and located in the core area corresponding to the first cluster A1 as the core set corresponding to the first cluster A1.

[0023] In this embodiment, the specific process of obtaining the out-of-cluster nearest neighbor point of the first cluster A1 is: all the first data points except the first cluster A1 are recorded as remaining data points, and the target out-of-cluster distance corresponding to each remaining data point is obtained. Among the target out-of-cluster distances corresponding to all the remaining data points, the remaining data point corresponding to the smallest target out-of-cluster distance is used as the out-of-cluster nearest neighbor point of the first cluster A1, and the method for obtaining the target out-of-cluster distance corresponding to the remaining data point is: for any remaining data point, in the first cluster A1, obtain a first data point closest to the remaining data point, and record it as the nearest neighbor data point of the remaining data point, and record the Euclidean distance between the nearest neighbor data point of the remaining data point and the remaining data point as the target out-of-cluster distance corresponding to the remaining data point.

[0024] In this embodiment, the specific process of obtaining the compactness representation value of each first core set according to all data points in each first core set and the initial clustering parameters is as follows: For any first core set B: first, obtain the number of first data points in the R neighborhood of each first data point in the first core set B, and record it as the neighborhood quantity representation value corresponding to the first data point; then record the cluster center of the first cluster cluster to which the first core set B belongs as the reference point of the first core set B; then obtain the Euclidean distance between each first data point in the first core set B and the reference point of the first core set B, and record it as the first distance corresponding to the first data point; then, according to the first distance of each first data point in the first core set B, obtain the feature distance weight value and mapping representation value corresponding to each first data point in the first core set B; then, according to the neighborhood quantity representation value, the corresponding feature distance weight value and mapping representation value corresponding to each first data point in the first core set B, obtain the first core set B. The comprehensive characterization value of each first data point in the first core set B; and for any first data point in the first core set B, the comprehensive characterization value of the first data point refers to the product of the normalized value of the neighborhood quantity characterization value corresponding to the first data point, the feature distance weight value and the mapping characterization value; then the product of the mean of the comprehensive characterization values ​​of all the first data points in the first core set B and the preset closeness weight value is recorded as the closeness characterization value of the first core set B, and in this embodiment, the normalized value of the neighborhood quantity characterization value corresponding to the first data point refers to the ratio of the neighborhood quantity characterization value corresponding to the first data point to the reference quantity characterization value of the first core set B, and the reference quantity characterization value of the first core set B refers to the maximum neighborhood quantity characterization value among the neighborhood quantity characterization values ​​corresponding to all the first data points in the first core set B.

[0025] In this embodiment, according to the first distance of each first data point in the first core set B, the specific process of obtaining the feature distance weight value and the mapping representation value corresponding to each first data point in the first core set B is: For any first data point in the first core set B: obtain the square value of the ratio of the first distance of the first data point to the reference distance of the first core set B, and record it as the characteristic ratio of the first data point, and record the inverse of the result obtained by adding the characteristic ratio of the first data point to the preset first constant as the characteristic distance weight value corresponding to the first data point, and the reference distance of the first core set B refers to the maximum first distance among the first distances of all the first data points in the first core set B; obtain all the first data points in the first core set B The mean of the first distance of the first data point is recorded as the first mean, the product of the first mean and the preset second constant is recorded as the target denominator value of the first core set B, the ratio of the square value of the first distance of the first data point to the target denominator value of the first core set B is recorded as the discrete representation value of the first data point, a negative correlation mapping is performed on the discrete representation value of the first data point, and the mapping result is recorded as the mapping representation value corresponding to the first data point, and the discrete representation value of the first data point is obtained based on the Gaussian kernel function, then the mapping representation value corresponding to the first data point is , and exp() is an exponential function with a constant e as the base, d is the first distance of the first data point, is the first mean, and c2 is the preset second constant.

[0026] In this embodiment, it is necessary to set the preset first constant, the preset second constant and the preset compactness weight value according to the actual situation. For example, in this embodiment, the preset first constant can be set to 1, the preset second constant can be set to 2, and the preset compactness weight value can be set to 1 or 2; and as another real-time method, the product of the mean of the comprehensive representation values ​​of all the first data points in the first core set B and the compactness coefficient corresponding to the first core set B can also be used as the compactness representation value of the first core set B. The role of the preset compactness weight value or the compactness coefficient is to amplify the weight of the high-density area, that is, to make the high-density area The compactness characterization value of the area is relatively large; and the process of obtaining the compactness coefficient corresponding to the first core set B is: record the set composed of the cluster centers of all the first clusters except the first cluster to which the first core set B belongs as a subset; in the subset, obtain the core set of the first cluster to which the cluster center closest to the reference point of the first core set B belongs, and record it as the nearest neighbor first core set corresponding to the first core set B, and take the ratio of the total number of the first data points in the first core set B to the total number of the first data points in its corresponding nearest neighbor core set as the compactness coefficient corresponding to the first core set B.

[0027] In addition, in this embodiment, the specific expression for obtaining the compactness representation value of the first core set B is:

[0028] Among them, C is the compactness representation value of the first core set B, is the preset compactness weight value, I0 is the number of the first data point in the first core set B, is the neighborhood quantity representation value corresponding to the i-th first data point in the first core set B, is the maximum neighborhood quantity representation value among the neighborhood quantity representation values ​​corresponding to all the first data points in the first core set B, and The main function is to normalize the neighborhood quantity representation value corresponding to the first data point. is the first distance of the ith first data point in the first core set B, is the reference distance of the first core set B, is the first mean, is the square value of the first distance of the ith first data point in the first core set B, is the feature distance weight value corresponding to the i-th first data point, is the mapping representation value corresponding to the i-th data point.

[0029] And when The bigger, The smaller and The smaller the value of C is, the larger the value of C is, and the larger the value of C is, the more densely distributed the data points in the first core set B are; on the contrary, when The smaller, The bigger and The larger it is, the smaller the value of C is. The smaller the value of C is, the less dense the distribution of data points in the first core set B is.

[0030] In this embodiment, the specific process of obtaining the intersection area between any two clusters is: for the first cluster K1 and the first cluster K2, the first cluster K1 and the first cluster K2 are not the same first cluster, obtain the minimum circumscribed circle containing the first cluster K1 and the minimum circumscribed circle containing the first cluster K2, and record the intersection area of ​​the minimum circumscribed circle containing the first cluster K1 and the minimum circumscribed circle containing the first cluster K2 as the intersection area between the first cluster K1 and the first cluster K2, and the method for obtaining the intersection area between any two second clusters is the same as the method for obtaining the intersection area between any two first clusters.

[0031] In addition, in specific applications, it is necessary to set a preset quantity threshold according to actual conditions. For example, in this embodiment, 2% of the total number of location data points on the navigation map can be used as the preset quantity threshold.

[0032] In this embodiment, the process of obtaining the compactness characterization value of the second set is as follows: first, the number of second data points in the R neighborhood of each second data point in the second set is obtained, and recorded as the neighborhood number characterization value corresponding to the corresponding second data point; then the average data point of all the second data points in the second set is used as the reference point of the second set, and the horizontal coordinate value of the reference point of the second set is the average value of the horizontal coordinate values ​​of all the second data points in the second set, and the vertical coordinate value of the reference point of the second set is the average value of the vertical coordinate values ​​of all the second data points in the second set; then, the relationship between each second data point in the second set and the reference point of the second set is obtained. The Euclidean distance between the two data points is calculated and recorded as the distance to be analyzed of the corresponding second data point; then, according to the distance to be analyzed of each second data point in the second set, the characteristic distance weight value and mapping characterization value corresponding to each second data point in the second set are obtained, and the method of obtaining the characteristic distance weight value and mapping characterization value corresponding to each second data point in the second set according to the distance to be analyzed of each second data point in the second set is the same as the method of obtaining the characteristic distance weight value and mapping characterization value corresponding to each first data point in the first core set B according to the first distance of each first data point in the first core set B, so it will not be described in detail. Then, the comprehensive characterization value of each second data point in the second set is obtained, and for any second data point in the second set, the comprehensive characterization value of the second data point is the product of the normalized value of the neighborhood quantity characterization value corresponding to the second data point, the characteristic distance weight value and the mapping characterization value; finally, the product of the mean of the comprehensive characterization values ​​of all second data points in the second set and the preset closeness weight value is recorded as the closeness characterization value of the second set.

[0033] In this embodiment, the initial clustering parameters are adjusted according to the compactness representation value of the second set and the first compactness threshold, and the specific process of obtaining the first adjusted clustering parameters is: First, the normalized value of the absolute value of the difference between the compactness characterization value of the second set and the first compactness threshold is recorded as the first characteristic difference value, and the normalization function norm() is used here to normalize the data that needs to be normalized; then, it is determined whether the first compactness threshold is less than the compactness characterization value of the second set. If so, the preset first adjustment coefficient is multiplied by the first characteristic difference value and then the result of adding the preset first constant is recorded as the first radius adjustment degree, and the preset first adjustment coefficient is multiplied by the nth power of the first characteristic difference value and then the inverse of the result obtained by adding the preset first constant is recorded as the first minimum point adjustment degree; and if it is determined that the first compactness threshold is not less than the compactness characterization value of the second set, the preset second adjustment coefficient is multiplied by the first characteristic difference value and then the result of adding the preset first constant is recorded as the first minimum point adjustment degree. The result of the preset first constant is recorded as the first radius adjustment degree, and the reciprocal of the result obtained by multiplying the preset second adjustment coefficient by the nth power of the first feature difference value and adding the preset first constant is recorded as the first minimum point adjustment degree, where n is the dimension of the position data point, and n is 2 in this embodiment; then the product of the first radius adjustment degree and the initial radius R is recorded as the first adjustment radius R1, and the result obtained by adding the preset first constant and the preset third adjustment coefficient is recorded as the target adjustment weight, and the result of multiplying the target adjustment weight, the first minimum point adjustment degree and the initial minimum point M is recorded as the first adjustment minimum point M1, and the first adjustment radius R1 and the first adjustment minimum point M1 both belong to the first adjustment clustering parameter, and the purpose of the above-mentioned target adjustment weight is mainly to further control the adjustment ratio.

[0034] In addition, when the first compactness threshold is less than the compactness representation value of the second set, the first radius adjustment degree is , the first minimum point adjustment level is The reciprocal of G1 is the first compactness threshold, H is the compactness representation value of the second set, is the preset first adjustment coefficient. When the first compactness threshold is less than the compactness representation value of the second set, the first radius adjustment degree is , the first minimum point adjustment level is The reciprocal of The second adjustment coefficient is preset; and in a specific application, the implementer needs to set the first adjustment coefficient, the second adjustment coefficient and the third adjustment coefficient according to the actual situation, and the first adjustment coefficient is required to be a negative number, and the second adjustment coefficient is a positive number. For example, in this embodiment, the first adjustment coefficient can be set to -0.2, the second adjustment coefficient can be set to 0.2, and the third adjustment coefficient can be set to 0.1.

[0035] And the above adjustment criteria are: when the first closeness threshold is less than the closeness representation value of the second set, it is considered that the closeness of the data points in the second set is greater than the closeness of the first core set obtained above, and in order to make the subsequent clustering results more reliable, the value of R should be reduced and the value of M should be increased; when the first closeness threshold is not less than the closeness representation value of the second set, it is considered that the closeness of the data points in the second set is not greater than the closeness of the first core set obtained above, and in order to make the subsequent clustering results more reliable, the value of R should be increased and the value of M should be decreased.

[0036] In this embodiment, according to the cluster center of each second cluster, the specific process of obtaining the core set corresponding to each second cluster is as follows: for any second cluster A2, among all the second data points except the second cluster A2, obtain a second data point closest to the second cluster A2 and record it as the out-cluster nearest neighbor of the second cluster A2; record the Euclidean distance between the cluster center of the second cluster A2 and the out-cluster nearest neighbor of the second cluster A2 as the out-cluster nearest neighbor distance of the second cluster A2; and take the cluster center of the second cluster A2 as the center and the second cluster A2 as the center. The area of ​​the circle with the outer neighbor distance of cluster 2 as the radius is recorded as the core area corresponding to the second cluster A2, and the set formed by all the second data points belonging to the second cluster A2 and located in the core area corresponding to the second cluster A2 is recorded as the core set corresponding to the second cluster A2; and the method of obtaining the second data point closest to the second cluster A2 among all the second data points except the second cluster A2 is the same as the method of obtaining the first data point closest to the first cluster A1 among all the first data points except the first cluster A1, so it will not be described in detail.

[0037] In this embodiment, the specific process of obtaining the compactness representation value of each second core set based on all the data points in each second core set and the first adjusted clustering parameter is as follows: for any second core set V: first, obtain the number of second data points in the R1 neighborhood of each second data point in the second core set V, and record it as the neighborhood quantity representation value corresponding to the corresponding second data point; then record the cluster center of the second cluster cluster to which the second core set B belongs as the reference point of the second core set V; then, record the Euclidean distance between each second data point in the second core set V and the reference point of the second core set V as the second distance of the corresponding second data point, and obtain the feature distance weight value and mapping representation value corresponding to each second data point in the second core set V based on the second distance of each second data point in the second core set V; obtain the comprehensive representation value of each second data point in the second core set V based on the neighborhood quantity representation value, feature distance weight value and mapping representation value corresponding to each second data point in the second core set V; finally The product of the mean of the comprehensive characterization values ​​of each second data in the second core set V and the preset closeness weight value is recorded as the closeness characterization value of the second core set V; and the method for obtaining the comprehensive characterization value of each second data point in the second core set V according to the neighborhood quantity characterization value, feature distance weight value and mapping characterization value corresponding to each second data point in the second core set V is the same as the method for obtaining the comprehensive characterization value of each first data point in the first core set B according to the neighborhood quantity characterization value, feature distance weight value and mapping characterization value corresponding to each first data point in the first core set B, so it will not be described in detail again. The method for obtaining the feature distance weight value and mapping characterization value corresponding to each second data point in the second core set V according to the second distance of each second data point in the second core set V is the same as the method for obtaining the feature distance weight value and mapping characterization value corresponding to each first data point in the first core set B according to the first distance of each first data point in the first core set B, so it will not be described in detail again.

[0038] In this embodiment, the process of obtaining the first tightness threshold and the first tightness threshold is: taking the average of the tightness characterization values ​​of all the first core sets as the first tightness threshold, and taking the average of the tightness characterization values ​​of all the second core sets as the second tightness threshold.

[0039] In this embodiment, the method of adjusting the first clustering parameter, continuing to cluster the third data point according to the adjusted clustering parameter, and obtaining the low-density set and the high-density set again according to the clustering result is the same as the above-mentioned method of adjusting the initial clustering parameter, continuing to cluster the second data point according to the adjusted clustering parameter, and continuing to obtain the low-density set and the high-density set according to the clustering result, so this embodiment will not be described in detail.

[0040] Therefore, this embodiment can obtain multiple low-density sets and multiple high-density sets through the above process. After obtaining the low-density set and the high-density set, the geometric transform coding compression algorithm is used to compress the data points in the obtained high-density set, and the compressed data is recorded as compressed data. The LZ compression algorithm is used to compress the data points in the obtained low-density set, and the compressed data is also recorded as compressed data. The geometric transform coding compression algorithm uses the spatial correlation and regularity of the data to improve the efficiency of subsequent data transmission through differential coding or morphological compression methods. The LZ compression algorithm is based on the replacement and pattern matching of repeated fragments to effectively handle sparse and scattered data, thereby improving the efficiency of subsequent data transmission, and the improvement in transmission efficiency can enable users to obtain more reasonable driving or walking routes when using navigation maps to plan driving or walking routes, and can also update road conditions more promptly to improve users' use of navigation maps; in addition, it should be noted that the above-mentioned compression method can not only improve data transmission efficiency and reduce cloud storage space, but also greatly improve the processing efficiency and access speed of map navigation data; and it should also be noted that the process of compressing data using geometric transform coding compression algorithm and LZ compression algorithm is a well-known technology, so the specific compression process will not be described in detail in this embodiment.

[0041] In summary, the present embodiment first obtains the location data points on the navigation map; then the location data points on the navigation map are all recorded as the first data points, and all the first data points are clustered using a clustering algorithm whose clustering parameters are the initial clustering parameters. The cluster clusters obtained after the clustering is completed are all recorded as the first cluster clusters, and according to the cluster centers of the first cluster clusters, the core sets corresponding to the first cluster clusters are obtained and recorded as the first core sets, and according to all the data points in the first core sets and the initial clustering parameters, the compactness representation values ​​of the first core sets are obtained; then, it is determined whether the compactness representation values ​​of the first core sets are greater than the first compactness threshold value, and if so, the corresponding first core set is recorded as a high-density set, otherwise, the corresponding first core set is recorded as a low-density set. Then, the intersection area between any two first clusters is obtained, and they are all recorded as the first intersection area, and it is determined whether the total number of the first data points in all the first intersection areas is not greater than the preset number threshold. If so, the set composed of all the first data points that do not belong to the first core set is recorded as a low-density set, otherwise, all the first data points that do not belong to the first core set are recorded as second data points, and the initial clustering parameters are adjusted, and the second data points are continued to be clustered according to the adjusted clustering parameters, and the low-density set and the high-density set are continued to be obtained according to the clustering results; finally, the data points in the high-density set are compressed by using the geometric transform coding compression algorithm to obtain compressed data, and the data points in the low-density set are compressed by using the LZ compression algorithm to obtain compressed data. In addition, this embodiment divides the data into low-density sets and high-density sets, and then compresses the low-density sets and the high-density sets using different compression algorithms, respectively, thereby improving the data compression effect and thus also improving the transmission efficiency. That is, the transmission efficiency of the compressed data can be subsequently improved when it is transmitted to the data analysis and processing module in the navigation map system, and the efficiency and rationality of the route planning of the navigation map system can also be improved, and more timely road condition updates can be ensured.

[0042] The embodiments described above are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, a person skilled in the art should understand that the technical solutions described in the aforementioned embodiments may still be modified, or some of the technical features may be replaced by equivalents. Such modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application, and should all be included in the protection scope of the present application.

Claims

1. A cloud storage-based map navigation data efficient processing system, comprising a processor and a memory, characterized in that: The processor executes the computer program stored in the memory to implement the following steps: Get location data points on the navigation map; The location data points on the navigation map are all recorded as first data points, and all the first data points are clustered using a clustering algorithm whose clustering parameters are initial clustering parameters, and the cluster clusters obtained after the clustering are completed are all recorded as first cluster clusters; according to the cluster center of each first cluster cluster, the core set corresponding to each first cluster cluster is obtained, and each is recorded as the first core set; according to all the data points in each first core set and the initial clustering parameter, the compactness representation value of each first core set is obtained; Determine whether the compactness representation value of each of the first core sets is greater than the first compactness threshold, if so, record the corresponding first core set as a high-density set, otherwise, record the corresponding first core set as a low-density set; Obtain the intersection area between any two first clusters, and record them as the first intersection area, determine whether the total number of first data points in all the first intersection areas is not greater than a preset number threshold, if so, record the set composed of all first data points that do not belong to the first core set as a low-density set, otherwise, record all first data points that do not belong to the first core set as second data points, and adjust the initial clustering parameters, continue to cluster the second data points according to the adjusted clustering parameters, and continue to obtain low-density sets and high-density sets according to the clustering results; The data points in the high-density set are compressed using a geometric transformation coding compression algorithm to obtain compressed data, and the data points in the low-density set are compressed using an LZ compression algorithm to obtain compressed data.

2. The efficient processing system for map navigation data based on cloud storage according to claim 1, characterized in that: The initial clustering parameters include an initial radius R and an initial minimum number of points M.

3. The efficient processing system for map navigation data based on cloud storage as claimed in claim 2, characterized in that: The method for obtaining the compactness representation value of the first core set includes: For any first core set B: obtain the number of first data points in the R neighborhood of each first data point in the first core set B, and record it as the neighborhood quantity representation value corresponding to the first data point; record the cluster center of the first cluster cluster to which the first core set B belongs as the reference point of the first core set B; record the Euclidean distance between each first data point in the first core set B and the reference point of the first core set B as the first distance of the corresponding first data point; according to the first distance of each first data point in the first core set B, obtain the feature distance weight value and mapping representation value corresponding to each first data point in the first core set B; For any first data point in the first core set B, the product of the normalized value of the neighborhood quantity representation value, the feature distance weight value and the mapping representation value corresponding to the first data point is recorded as the comprehensive representation value of the first data point; The product of the average of the comprehensive representation values ​​of all the first data points in the first core set B and the preset compactness weight value is recorded as the compactness representation value of the first core set B.

4. The cloud storage-based map navigation data efficient processing system according to claim 3, characterized in that: The method of obtaining a feature distance weight value and a mapping representation value corresponding to each first data point in the first core set B according to the first distance of each first data point in the first core set B includes: Among the first distances of all the first data points in the first core set B, the maximum first distance is selected as the reference distance of the first core set B; the square value of the ratio of the first distance of each first data point in the first core set B to the reference distance of the first core set B is recorded as the characteristic ratio of the corresponding first data point; The reciprocal of the result obtained by adding the feature ratio of each first data point in the first core set B to the preset first constant is recorded as the feature distance weight value corresponding to the first data point; Obtain the product of the mean value of the first distances of all the first data points in the first core set B and a preset second constant, and record it as the target denominator value of the first core set B; record the ratio of the square value of the first distance of each first data point in the first core set B to the target denominator value of the first core set B as the discrete representation value corresponding to the first data point, and record the negative correlation mapping value of the discrete representation value of the first data point as the mapping representation value corresponding to the first data point.

5. The efficient processing system for map navigation data based on cloud storage according to claim 4, characterized in that: The method of adjusting the initial clustering parameters, continuing to cluster the second data points according to the adjusted clustering parameters, and continuing to obtain a low-density set and a high-density set according to the clustering results includes: Recording a set consisting of all second data points in the navigation map as a second set, obtaining a compactness representation value of the second set, and adjusting the initial clustering parameter according to the compactness representation value of the second set and a first compactness threshold to obtain a first adjusted clustering parameter; All the second data points are clustered using a clustering algorithm whose clustering parameters are the first adjusted clustering parameters, and the cluster clusters obtained after the clustering are all recorded as second cluster clusters; according to the cluster centers of each second cluster cluster, the core sets corresponding to each second cluster cluster are obtained, and are all recorded as second core sets; according to all the data points in each second core set and the first adjusted clustering parameters, the compactness characterization value of each second core set is obtained; the method for obtaining the compactness characterization value of the second core set is the same as the method for obtaining the compactness characterization value of the first core set; Determine whether the compactness representation value of each second core set is greater than the second compactness threshold, if so, record the corresponding second core set as a high-density set, otherwise, record the corresponding second core set as a low-density set; Obtain the intersection area between any two second clusters and record them as the second intersection area, determine whether the total number of second data points in all the second intersection areas is not greater than a preset threshold value, if so, record the set consisting of all second data points that do not belong to the second core set as a low-density set, otherwise, record all second data points that do not belong to the second core set as third data points, and adjust the first adjustment clustering parameters, continue to cluster the third data points according to the adjusted clustering parameters, and obtain low-density sets and high-density sets again according to the clustering results, and so on, until all the low-density sets and high-density sets obtained contain all the location data points on the navigation map and stop.

6. The efficient processing system for map navigation data based on cloud storage according to claim 5, characterized in that: The method for obtaining the core set corresponding to the cluster includes: For any first cluster A1, among all the first data points except the first cluster A1, obtain a first data point closest to the first cluster A1, and record it as the out-cluster nearest neighbor point of the first cluster A1, record the Euclidean distance between the cluster center of the first cluster A1 and the out-cluster nearest neighbor point of the first cluster A1 as the out-cluster nearest neighbor distance of the first cluster A1, record the area of ​​a circle with the cluster center of the first cluster A1 as the center and the out-cluster nearest neighbor distance of the first cluster A1 as the radius as the core area corresponding to the first cluster A1, and record the set formed by all the first data points belonging to the first cluster A1 and located in the core area corresponding to the first cluster A1 as the core set corresponding to the first cluster A1; For any second cluster A2, among all the second data points except the second cluster A2, obtain a second data point closest to the second cluster A2, and record it as the out-cluster nearest neighbor point of the second cluster A2, record the Euclidean distance between the cluster center of the second cluster A2 and the out-cluster nearest neighbor point of the second cluster A2 as the out-cluster nearest neighbor distance of the second cluster A2, record the area of ​​the circle with the cluster center of the second cluster A2 as the center and the out-cluster nearest neighbor distance of the second cluster A2 as the radius as the core area corresponding to the second cluster A2, and record the set formed by all the second data points belonging to the second cluster A2 and located in the core area corresponding to the second cluster A2 as the core set corresponding to the second cluster A2.

7. The efficient processing system for map navigation data based on cloud storage according to claim 5, characterized in that: The first compactness threshold is the average of the compactness representation values ​​of all the first core sets, and the second compactness threshold is the average of the compactness representation values ​​of all the second core sets.

8. The efficient processing system for map navigation data based on cloud storage as claimed in claim 5, characterized in that: The method for obtaining the compactness representation value of the second set includes: Obtaining the number of second data points in the R neighborhood of each second data point in the second set, and recording it as a neighborhood quantity representation value corresponding to the second data point; The average data point of all data points in the second set is used as the reference point of the second set; the Euclidean distance between each second data point in the second set and the reference point of the second set is recorded as the distance to be analyzed of the corresponding second data point; according to the distance to be analyzed of each second data point in the second set, the characteristic distance weight value and mapping characterization value corresponding to each second data point in the second set are obtained; the method for obtaining the characteristic distance weight value and mapping characterization value corresponding to each second data point in the second set according to the distance to be analyzed of each second data point in the second set is the same as the method for obtaining the characteristic distance weight value and mapping characterization value corresponding to each first data point in the first core set B according to the first distance of each first data point in the first core set B; For any second data point in the second set, the product of the normalized value of the neighborhood quantity representation value, the feature distance weight value and the mapping representation value corresponding to the second data point is recorded as the comprehensive representation value of the second data point; The product of the average of the comprehensive representation values ​​of all the second data points in the second set and the preset compactness weight value is recorded as the compactness representation value of the second set.

9. The efficient processing system for map navigation data based on cloud storage according to claim 5, characterized in that: The method of adjusting the initial clustering parameters according to the compactness representation value of the second set and the first compactness threshold to obtain a first clustering parameter adjustment includes: Recording a normalized value of the absolute value of the difference between the compactness representation value of the second set and the first compactness threshold as a first feature difference value; If the first closeness threshold is less than the closeness representation value of the second set, the result of multiplying the first characteristic difference value by the preset first adjustment coefficient and adding the preset first constant is recorded as the first radius adjustment degree, and the reciprocal of the result of multiplying the first characteristic difference value by the preset first adjustment coefficient and adding the preset first constant is recorded as the first minimum point adjustment degree; otherwise, the result of multiplying the first characteristic difference value by the preset second adjustment coefficient and adding the preset first constant is recorded as the first radius adjustment degree, and the reciprocal of the result of multiplying the first characteristic difference value by the preset second adjustment coefficient and adding the preset first constant is recorded as the first minimum point adjustment degree, where n is the dimension of the position data point; The product of the first radius adjustment degree and the initial radius R is recorded as the first adjustment radius R1, the result obtained by adding the preset first constant and the preset third adjustment coefficient is recorded as the target adjustment weight, and the result of multiplying the target adjustment weight, the first minimum point adjustment degree and the initial minimum point M is recorded as the first adjustment minimum point M1. The first adjustment radius R1 and the first adjustment minimum point M1 both belong to the first adjustment clustering parameter.

10. The cloud storage-based map navigation data efficient processing system according to claim 5, characterized in that: The method for obtaining the intersection area includes: For the first cluster K1 and the first cluster K2, the intersection area between the minimum circumscribed circle containing the first cluster K1 and the minimum circumscribed circle containing the first cluster K2 is recorded as the intersection area between the first cluster K1 and the first cluster K2; the method for obtaining the intersection area between any two second clusters is the same as the method for obtaining the intersection area between any two first clusters.

Images