CN114613124A - Traffic information processing method, device, terminal and computer-readable storage medium - Google Patents

Abstract

The present invention is applicable to the field of intelligent transportation, and provides a traffic information processing method, device, terminal and computer-readable storage medium. The method includes: acquiring traffic point information; converting the traffic point information into traffic track information; according to the traffic track information Determine the adjacency relationship between any two adjacent traffic points in the traffic trajectory and the number of traffic passes; based on the information of the traffic point and the adjacency relationship between any two adjacent traffic points and the number of passes, use the spectral clustering method to obtain a high-dimensional map of the traffic network Feature matrix; extract category information of traffic points from high-dimensional representation; classify and store traffic points based on category information to form a database. In this technical solution, the number of track communications is used to measure the similarity between traffic points, and the spectral clustering algorithm is used to characterize the relationship between spatiotemporal data from the perspective of a graph network, so that the clustering results of traffic points are more realistic. of the transportation network.

Description

Traffic information processing method, device, terminal and computer-readable storage medium

technical field

The present invention belongs to the field of intelligent traffic, and in particular relates to a traffic information processing method, device, device and computer-readable storage medium.

Background technique

In the field of urban computing and intelligent transportation, plane clustering methods are widely used, such as the analysis and prediction of traffic passenger flow trends, and the division of public transportation areas. By using the clustering method, the plane of the city can be divided into different classes, and each class may represent different roads, communities, etc., and can further excavate business districts, residences, and congested road sections.

In recent years, with the development of urban big data and intelligent transportation technology, more and more information can be mined from the data, which further provides support for different applications, such as business district mining, precise advertising, traffic capacity analysis, and passenger flow forecasting. and many more. In order to facilitate analysis, such applications require a basic premise, that is, the city floor plan needs to be discretized, that is, the city plane is divided into different small blocks, so that each small block can be labeled with different labels. Support for further analysis.

Usually the data we collect are GPS data reported by traffic participants, such as buses, online car-hailing, taxis, subways, private cars, etc. These GPS data represent different roads that can pass in the city. How to combine these The division of GPS points into different clusters is a fairly basic and important problem.

For the clustering of the spatial plane, there are currently many methods, such as K-Means, DBSCAN, and a large number of variants based on these two. By calculating the Euclidean distance between GPS points, combining these two methods can be used to delineate GPS points. For different classes, the space plane is divided into different blocks.

The main problem with the above-mentioned prior art is that they use the Euclidean distance as a measurement index, that is, the distance relationship in space. But in real scenes, objects (such as vehicles or people) usually move in a spatial network, that is, in a network-structured manner, rather than in Euclidean plane space. For example, vehicles are often moving along city roads, and although two roads may be close in space, it is virtually impossible for a vehicle to move from one road across the barrier to the other. In K-Means or DBSCAN, which measures the relationship by Euclidean distance, the GPS reported by the vehicles on such two roads will be considered as one type, but this is obviously not in line with the actual situation. That is to say, using spatial distance for clustering will classify adjacent points that are completely impossible to intersect into the same class.

SUMMARY OF THE INVENTION

In view of this, the embodiments of the present invention provide a traffic information processing method, device, terminal, and computer-readable storage medium, so as to solve the problem that the use of Euclidean distance in traditional clustering algorithms results in inability to represent complex urban traffic networks during traffic information processing. The problem.

A first aspect of the embodiments of the present invention provides a method for processing traffic information, including:

Get information about traffic points;

converting the information of the traffic point into traffic track information;

Determine the adjacency relationship between any two adjacent traffic points in the traffic track and the number of track passes according to the traffic track information;

Based on the information of the traffic point and the adjacency relationship between any two adjacent traffic points and the number of passes, the method of spectral clustering is used to obtain a high-dimensional representation of the traffic network composed of the traffic points;

extracting category information of the traffic point from the high-dimensional representation;

Based on the category information, the information of the traffic point is classified and stored to form a database.

A second aspect of the embodiments of the present invention provides a device for processing traffic information, including:

The acquisition unit is used to acquire the information of the traffic point;

a conversion unit, configured to convert the information of the traffic point into traffic track information;

a determining unit, configured to determine the adjacency relationship between any two adjacent traffic points in the traffic track and the number of track passes according to the traffic track information;

a computing unit, configured to obtain a high-dimensional representation of the traffic network composed of the traffic points by using the method of spectral clustering based on the information of the traffic points and the adjacency relationship between any two adjacent traffic points and the number of passes;

an extraction unit, configured to extract the category information of the traffic point from the high-dimensional representation;

The storage unit is configured to classify and store the information of the traffic point based on the category information to form a database.

A third aspect of the embodiments of the present invention provides a terminal, including a memory, a processor, and a computer program stored in the memory and executable on the processor, which is implemented when the processor executes the computer program the steps of the above method.

A fourth aspect of the embodiments of the present invention provides a computer-readable storage medium, where a computer program is stored in the computer-readable storage medium, wherein when the computer program is executed by a processor, the steps of the above method are implemented.

The beneficial effects that the embodiment of the present invention has compared with the prior art are:

In this technical solution, the number of track communications is used to measure the similarity between traffic points, and the spectral clustering algorithm is used to characterize the relationship between spatiotemporal data from the perspective of a graph network, so that the clustering results of traffic points are more consistent with actual traffic network conditions.

Description of drawings

In order to illustrate the technical solutions in the embodiments of the present invention more clearly, the following briefly introduces the accompanying drawings that need to be used in the description of the embodiments or the prior art. Obviously, the drawings in the following description are only for the present invention. In some embodiments, for those of ordinary skill in the art, other drawings can also be obtained according to these drawings without any creative effort.

Fig. 1 is a flow chart of the first embodiment of the traffic information processing method of the present invention;

FIG. 2 is a schematic diagram of the refinement flow of S13 in the first embodiment of the traffic information processing method of the present invention;

3 is a schematic diagram of a refinement flow of S14 in the first embodiment of the traffic information processing method of the present invention;

4 is a schematic structural diagram of a distributed database system in the first embodiment of the traffic information processing method of the present invention;

Figure 5 is the effect diagram of different clustering algorithms;

6 is a flowchart of a second embodiment of the traffic information processing method of the present invention;

Figure 7 is a data query delay test chart;

FIG. 8 is a schematic structural diagram of the first embodiment of the traffic information processing device of the present invention;

FIG. 9 is a schematic structural diagram of a first embodiment of a terminal of the present invention.

Detailed ways

In the following description, for the purpose of illustration rather than limitation, specific details such as specific system structures and technologies are set forth in order to provide a thorough understanding of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the present invention may be practiced in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, devices, circuits, and methods are omitted so as not to obscure the description of the present invention with unnecessary detail.

In order to illustrate the technical solutions of the present invention, the following specific embodiments are used for description.

In the embodiment of the traffic information processing method of the present invention, the information of the traffic point is obtained by converting the information of the traffic point into the traffic track information, and the adjacency relationship between any two adjacent traffic points in the traffic track is determined according to the traffic track information. As well as the number of track passes, based on the information of traffic points and the adjacency relationship between any two adjacent traffic points and the number of passes, the high-dimensional graph feature matrix of the traffic network composed of traffic points is obtained by using the method of spectral clustering. And based on the category information of the traffic points extracted from the feature matrix of the high-dimensional graph, the traffic points are classified and stored to form a database.

In the embodiment of the present invention, the number of track communications is used to measure the similarity between traffic points, and the spectral clustering algorithm is used to characterize the relationship between spatiotemporal data from the perspective of a graph network, so that the clustering of traffic points The results are more in line with the actual traffic network situation.

FIG. 1 is a flowchart of the first embodiment of the traffic information processing method of the present invention. As shown in FIG. 1 , the traffic information processing method includes the following steps:

S11, acquiring information of the traffic point.

Among them, the information of the traffic point includes GPS data reported by traffic participants (such as public transport, online car-hailing, taxi, subway, private car, etc.). Each piece of GPS data can be represented as (participant ID, reporting time time, longitude longtitude, latitude latitude). The acquisition method may be direct receiving, or may be reading historical GPS data stored in the storage medium from a local or other storage medium.

S12: Convert the traffic point information into traffic track information.

In this embodiment, the traffic trajectory information includes a traffic trajectory sequence. Specifically, the GPS data reported by the same participant is grouped into one group according to the participant ID of the GPS data, and sorted according to the reporting time of the GPS data to obtain each traffic participant ID. The traffic trajectory sequence of the person.

In other embodiments of the method of the present invention, the GPS data may also be classified according to other parameters in the GPS data to obtain corresponding traffic track information.

S13: Determine the adjacency relationship between any two adjacent traffic points in the traffic track and the number of times of the track passing according to the traffic track information.

In this implementation, as shown in FIG. 2 , step S13 specifically includes the following sub-steps:

S131, splitting the traffic trajectory information into adjacent traffic point pairs;

S132, determine the adjacency relationship of any two adjacent traffic points according to the adjacent traffic point pair;

S133 , determining the number of track passes of any two adjacent traffic points according to the adjacent traffic point pair.

Specifically, in sub-step S131, the longitude and latitude of the GPS data can be encoded into a number lx by the method of geohash, that is, lx represents a specific traffic point, that is, a pair of longitude and latitude. Assuming that x1, x2, x3 and x4 are four traffic points adjacent one by one, the trajectory sequence can be expressed as (i,lx1,t1),(i,lx2,t2),(i,lx3,t3),(i , lx4, t4), where i is the ID of the traffic participant reporting the corresponding GPS data, lx1-lx4 correspond to the numbers obtained by geohash coding of the longitude and latitude of the traffic point x1-x4, and t1-t4 are the corresponding GPS data reporting time . The corresponding split-out adjacent traffic point pairs are denoted as (lx1,lx2), (lx2,lx3) and (lx3,lx4). Through the above method, each trajectory sequence is split into a set (la, lb) of adjacent traffic point pairs, where a and b represent any two adjacent traffic points.

In sub-step S132, each GPS point is regarded as a vertex, then there is an adjoining edge between the vertexes la and lb, that is, an adjacency relationship e(a, b).

In sub-step S133, since it is an undirected graph, the number of (la, lb) and (lb, la) appearing in the set of adjacent traffic point pairs can be counted, and the number of track passes of any two adjacent traffic points can be obtained. w(a,b), w(a,b) is expressed as:

w(a,b)=w'(a,b)+w'(b,a)

S14 , based on the information of the traffic points and the adjacency relationship between any two adjacent traffic points and the number of passes, a spectral clustering method is used to obtain a high-dimensional graph feature matrix of the traffic network composed of the traffic points.

In this embodiment, as shown in FIG. 3 , step S14 specifically includes the following sub-steps:

S141, determine the adjacency matrix with the number of times of the trajectory passing of any two adjacent traffic points as the weight value of the adjacency relationship of the corresponding two adjacent traffic points;

S142, according to the information of the traffic point and the adjacency relationship of any two adjacent traffic points and the adjacency matrix, obtain a weighted undirected adjacency graph of the traffic network composed of the traffic points;

S143, according to the weighted undirected adjacency graph, calculate a high-dimensional graph feature matrix of the traffic network composed of traffic points.

Specifically, in sub-step S141, set the total weight of the adjacency relationship e(a,b) to w(a,b), that is, the similarity between any two adjacent traffic points, so that an adjacency can be obtained. matrix W.

In sub-step S142, according to the data V of any GPS point and the adjacency relationship E of any two adjacent traffic points and the adjacency matrix W (the adjacency matrix indicates whether there is a connection between all vertices, if there is a connection, then the connection edge weight is value on the corresponding matrix) to obtain a weighted undirected adjacency graph G(V,E,W) of the traffic network composed of these GPS points.

In sub-step S143, according to the weighted undirected adjacency graph G(V, E, W), the degree (degree) di corresponding to each traffic point can be calculated:

After calculating the degrees corresponding to all traffic points, the degree matrix Deg can be obtained.

Based on the degree matrix Deg and the adjacency matrix W, the unregularized Laplacian matrix Lap of the weighted undirected adjacency graph G is calculated by the following formula:

Lap=Deg-W

After that, the eigenvectors and eigenvalues of the Laplacian matrix Lap are solved. In order to reduce the amount of computation and avoid the appearance of a large number of meaningless clusters, the TopK algorithm will be used for the feature vector _Λn :

Λ _n ←solve|Lap-λI|=0 extract the eigenvalues(λ ₁ , λ ₂ ,...,λ _n );

Λ _k ←TopK(Λ _n );

Specifically, the Plasma matrix Lap is n*n, and n eigenvalues (n is the number of vertices) are obtained by taking the eigenvalues above, and each eigenvalue corresponds to an eigenvector of length n. Then take topk (the largest k eigenvalues) according to the size of the eigenvalues, and arrange the eigenvectors corresponding to the first k eigenvalues in order to obtain a new matrix of k*n, which is the traffic composed of these GPS points. The high-dimensional graph feature matrix of the network.

S15, the category information of the traffic point is extracted from the high-dimensional graph feature matrix.

In this embodiment, clustering is performed on the feature matrix of the high-dimensional map, and the category information of different traffic points can be extracted, and the category information of the extracted traffic points includes the classification table of GPS points (longitude, latitude, category number) .

S16, classify and store the information of the traffic point based on the category information to form a database.

In this embodiment, after the classification table of GPS points is obtained, the original GPS data is stored in the database according to the classification table, and then the corresponding GPS data can be found out by using the category number clusterID for further analysis.

Specifically, a distributed database can be used to convert and store the original GPS data. For example, using HBase+Phoenix (Phoenix is a plug-in of HBase, which can provide HBase with a query interface similar to SQL, through which you can directly use SQL to query HBase data) The way to store data, the implementation of the clustering algorithm is based on Spark-SQL framework, so the architecture of the whole system is shown in Figure 4.

The GPS data RawData reported by traffic participants is classified by the above-mentioned algorithm running in the Analysis Engine. Analysis Engine is a big data framework and program that programs run, and Spark-SQL can be used to implement the above algorithms. The server node Master is the server node used by spark to manage the cluster, and the program is submitted through the master. The server node worker is the server node that runs the algorithm in a distributed manner. The program submitted to the master is divided and sent to different workers to run. After the operation is completed, the result is sent back to the master.

The client Query can run on the client's terminal device, such as the client's computer, and the data stored in HBASE can be queried by installing the SQL client.

Among them, Spark-SQL implements the above clustering algorithm, and stores the GPS data in HBase according to the generated classification table. HBase will use multiple servers (ie Region Server in the figure) as the storage medium, that is, the GPS data data is scattered on different servers. Specifically, by placing the ClusterID at the beginning of the row key Rowkey (HBase uses Rowkey to query data), and then placing the data whose ClusterID is c1 (Rowkey starts with c1) on the first server Region Server1, and the ClusterID is c2. Put it into the second server Region Server2, and so on, to realize that different types of GPS data are put into different servers.

In addition, in order to optimize the query efficiency, the RowKey of HBase can be designed with an inverted index:

RowKey=ClusterID+MaxValue-Timestamp+ID

Among them, ClusterID is the classification ID of the cluster, and MaxValue is the maximum value of time. The inverted index is realized by using the difference between it and the actual time, that is, the latest data is placed in the upper layer of the database, and the query efficiency is higher. In general, data is not evenly distributed in space, so using ClusterID as the prefix of RowKey will inevitably lead to hot issues. To solve this problem and make the computation parallelize, RowKey is distributed as widely as possible. For example, you can use Salted Tables, a function of Phoenix, to specify how many servers (Region Servers) there are. It will transparently add a Hash symbol in front of Rowkey, and then according to the above, the data of Rowkey at the beginning of different symbols will be placed To different servers, evenly put the data on different servers. Because the second string of rowkey is ClusterID, the data of the same ClusterID will be put on one server as much as possible.

For the taxi trajectory in Beijing, (a) KMeans method, (b) DBSCAN method and (c) the method of the first embodiment of the present invention (ie Spectral in FIG. 5 ) are used to classify GPS data, and the classification results are shown in the figure 5, it can be seen that clustering using (a) KMeans method and (b) DBSCAN method usually divides the space into relatively regular blocks, which are often inconsistent with the actual situation. Using (c) the method of the first embodiment of the present invention for clustering will use the form of a graph network to represent the space, and the different categories divided are more in line with the distribution of real tracks or roads.

In the first embodiment of the traffic information processing method of the present invention, the adjacency relationship between different traffic points is represented by a graph, and then the adjacency graph is used to perform cluster analysis to divide the urban traffic space plane. Aiming at the problem of measuring the correlation between different traffic points, the statistics of historical traffic data are used to analyze the number of track passes between two adjacent traffic points. The distance measurement error is more robust.

6 is a flowchart of a second embodiment of the traffic information processing method of the present invention. In this embodiment, the method of the present invention includes:

S61, obtain information of the traffic point;

S62, converting the traffic point information into traffic trajectory information;

S63, determine the adjacency relationship between any two adjacent traffic points in the traffic track and the number of track passes according to the traffic track information;

S64, based on the information of the traffic point and the adjacency relationship between any two adjacent traffic points and the number of passes, use the method of spectral clustering to obtain a high-dimensional graph feature matrix of the traffic network composed of the traffic points;

S65, extract the category information of the traffic point from the high-dimensional map feature matrix;

S66, classify and store the traffic point information based on the category information to form a database;

S67, receiving a request for obtaining information of a traffic point of a specific category;

S68, based on the category information in the request, obtain information of traffic points of the corresponding category from the database to respond.

In this embodiment, steps S61-S66 correspond to the same as S11-S16 in the first embodiment of the method, and are not repeated here.

Referring to FIG. 4 , in step S67 , a request for acquiring information of the traffic point is received from the client Query, and the request includes category information of the information of the traffic point. In step S68, based on the category information, the information of the traffic point of the corresponding category is obtained from the database to respond. For example, the information of the corresponding traffic point is sent to the corresponding client directly or after preset processing.

Due to the use of HBase+Phoenix to store data, Phoenix can provide a SQL-like query interface for HBase, through which you can directly use SQL to query HBase data. When querying, the SQL statement will be parsed into parallel statements by Phoenix. Data is queried in parallel on the storage medium to speed up efficiency and can be applied to millisecond-level real-time queries.

As shown in Fig. 7, the test is carried out on the 4.3 billion GPS data set. As the input scale of the query increases, the query delay increases approximately linearly. In most cases, the delay within 4s can be achieved, satisfying the real-time query requirements.

It should be understood that the size of the sequence numbers of the steps in the above embodiments does not mean the sequence of execution, and the execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

In the embodiment of the present invention, a traffic information processing apparatus is also provided, and each unit included in the traffic information processing apparatus is used to execute each step in the embodiment corresponding to FIG. 1 . For details, please refer to the relevant descriptions in the embodiments corresponding to FIGS. 1-5 . FIG. 8 shows a schematic structural diagram of the first embodiment of the traffic information processing apparatus 800 of the present invention, including:

an obtaining unit 81, used for obtaining the information of the traffic point;

A conversion unit 82, configured to convert the information of the traffic point into traffic track information;

A determination unit 83, configured to determine the adjacency relationship between any two adjacent traffic points in the traffic track and the number of track passes according to the traffic track information;

The computing unit 84 is used to obtain the high-dimensional map feature matrix of the traffic network formed by the traffic points by using the method of spectral clustering based on the information of the traffic points and the adjacency relationship of any two adjacent traffic points and the number of passes;

The extraction unit 85 is used for extracting the category information of the traffic point from the high-dimensional representation;

The storage unit 86 is configured to classify and store the information of the traffic point based on the category information to form a database.

Wherein, the determining unit 83 includes:

a first determining module 831, configured to split the traffic trajectory information into adjacent traffic point pairs;

The second determining module 832 is configured to determine the adjacency relationship between any two adjacent traffic points according to the adjacent traffic point pair;

The third determining module 833 is configured to determine the number of track passes of any two adjacent traffic points according to the adjacent traffic point pair.

Wherein, the computing unit 84 includes:

The adjacency matrix calculation module 841 is used to determine the adjacency matrix by taking the number of tracks of any two adjoining traffic points as the weight value of the adjacency relationship of the corresponding two adjoining traffic points;

The adjacency graph calculation module 842 is used to obtain the weighted undirected adjacency graph of the traffic network composed of the traffic points according to the information of the traffic points and the adjacency relationship between any two adjacent traffic points and the adjacency matrix;

The high-dimensional representation calculation module 843 is configured to calculate a high-dimensional graph feature matrix of a traffic network composed of traffic points according to the weighted undirected adjacency graph.

Further, the high-dimensional representation calculation module 843 includes the following sub-modules (not shown in the figure):

The first sub-module is used to calculate the degrees corresponding to all traffic points according to the weighted undirected adjacency graph to obtain a degree matrix;

The second submodule is used to calculate the Laplacian matrix of the weighted undirected adjacency graph based on the degree matrix and the adjacency matrix;

The third submodule is used to solve the eigenvectors and eigenvalues of the Laplace matrix;

The fourth sub-module is used to obtain a high-dimensional graph feature matrix of a traffic network composed of traffic points by using the TopK algorithm according to the eigenvectors and eigenvalues of the Laplace matrix.

In the second embodiment of the traffic information processing device of the present invention, based on FIG. 8 , the device further includes a receiving unit and a response unit for performing the corresponding steps in the embodiment corresponding to FIG. 6 . For details, please refer to the corresponding Relevant descriptions in the examples.

Wherein, the receiving unit is configured to receive a request for acquiring information of a traffic point of a specific category.

The response unit is used for the category information in the base request, and obtains the traffic point information of the corresponding category from the database to respond.

The present invention also provides a terminal. As shown in FIG. 9 , the terminal 100 includes: a processor 101 , a memory 102 , and a computer program 103 stored in the memory 102 and running on the processor 101 . When the processor 101 executes the computer program 103, the steps in each embodiment of the traffic information processing method described above are implemented. Alternatively, when the processor 101 executes the computer program 103, the functions of the units/modules/sub-modules in the foregoing apparatus embodiments are implemented.

Exemplarily, the computer program 103 may be divided into one or more units/modules/submodules that are stored in the memory 102 and executed by the processor 101 to complete the this invention. The one or more units/modules/submodules may be a series of computer program instruction segments capable of performing specific functions, and the instruction segments are used to describe the execution process of the computer program 103 in the above-mentioned traffic information processing device/terminal 100 . For example, the computer program 62 can be divided into an acquisition module, an execution module, and a generation module (modules in a virtual device), and the specific functions of each module are as follows:

Obtain the information of traffic points; convert the information of traffic points into traffic trajectory information; determine the adjacency relationship between any two adjacent traffic points in the traffic trajectory and the number of track passes based on the traffic trajectory information; based on the information of traffic points and any two adjacencies The adjacency relationship and the number of traffic points of the traffic points are obtained by using the spectral clustering method to obtain the high-dimensional map feature matrix of the traffic network composed of traffic points; the category information of the traffic points is extracted from the high-dimensional map feature matrix; based on the category The information is classified and stored for traffic points to form a database.

The terminal 100 may be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server. The terminal 100 may include, but is not limited to, a processor 101 and a memory 102 . Those skilled in the art can understand that FIG. 9 is only an example of the terminal 100, and does not constitute a limitation to the terminal 100, and may include more or less components than the one shown, or combine some components, or different components, such as The terminal 100 may also include input and output devices, network access devices, buses, and the like.

The processor 101 may be a central processing unit (Central Processing Unit, CPU), other general-purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC), an off-the-shelf processor Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.

The memory 102 may be an internal storage unit of the terminal 100 , such as a hard disk or a memory of the terminal 100 . The memory 102 may also be an external storage device of the terminal 100, such as a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, and a flash memory card (Flash Card) equipped on the terminal 100. Wait. Further, the memory 102 may also include both an internal storage unit of the terminal 100 and an external storage device. The memory 102 is used to store the computer program 103 and other programs and data required by the terminal 100 . The memory 102 may also be used to temporarily store data that has been or will be output.

The present invention also provides a computer-readable storage medium, where the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, implements the steps in any of the embodiments of the traffic information processing method.

Those skilled in the art can clearly understand that, for the convenience and simplicity of description, only the division of the above-mentioned functional units and modules is used as an example. Module completion, that is, dividing the internal structure of the device into different functional units or modules to complete all or part of the functions described above. Each functional unit and module in the embodiment may be integrated in one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit, and the above-mentioned integrated units may adopt hardware. It can also be realized in the form of software functional units. In addition, the specific names of the functional units and modules are only for the convenience of distinguishing from each other, and are not used to limit the protection scope of the present application. For the specific working process of the units and modules in the above-mentioned system, reference may be made to the corresponding process in the foregoing method embodiments, which will not be repeated here.

In the foregoing embodiments, the description of each embodiment has its own emphasis. For parts that are not described or described in detail in a certain embodiment, reference may be made to the relevant descriptions of other embodiments.

Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans may implement the described functionality using different methods for each particular application, but such implementations should not be considered beyond the scope of the present invention.

In the embodiments provided by the present invention, it should be understood that the disclosed apparatus/terminal device and method may be implemented in other manners. For example, the apparatus/terminal device embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units. Or components may be combined or may be integrated into another system, or some features may be omitted, or not implemented. On the other hand, the shown or discussed mutual coupling or direct coupling or communication connection may be through some interfaces, indirect coupling or communication connection of devices or units, and may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution in this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically alone, or two or more units may be integrated into one unit. The above-mentioned integrated units may be implemented in the form of hardware, or may be implemented in the form of software functional units.

The integrated modules/units, if implemented in the form of software functional units and sold or used as independent products, may be stored in a computer-readable storage medium. Based on this understanding, the present invention can implement all or part of the processes in the methods of the above embodiments, and can also be completed by instructing relevant hardware through a computer program, and the computer program can be stored in a computer-readable storage medium. When the program is executed by the processor, the steps of the foregoing method embodiments can be implemented. . Wherein, the computer program includes computer program code, and the computer program code may be in the form of source code, object code, executable file or some intermediate form, and the like. The computer-readable medium may include: any entity or device capable of carrying the computer program code, a recording medium, a U disk, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM, Read-Only Memory) , Random Access Memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer-readable media may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, the computer-readable media Electric carrier signals and telecommunication signals are not included.

The above-mentioned embodiments are only used to illustrate the technical solutions of the present invention, but not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: it is still possible to implement the foregoing implementations. The technical solutions described in the examples are modified, or some technical features thereof are equivalently replaced; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the within the protection scope of the present invention.

Claims (10)

1. A method for processing traffic information, wherein the method comprises: Get information about traffic points; converting the information of the traffic point into traffic track information; Determine the adjacency relationship between any two adjacent traffic points in the traffic track and the number of track passes according to the traffic track information; Based on the information of the traffic point and the adjacency relationship between any two adjacent traffic points and the number of passes, the method of spectral clustering is used to obtain a high-dimensional representation of the traffic network composed of the traffic points; extracting category information of the traffic point from the high-dimensional representation; Based on the category information, the information of the traffic point is classified and stored to form a database. 2. The method according to claim 1, wherein the information of the traffic point comprises GPS data, and the converting the information of the traffic point into traffic track information comprises: The GPS data reported by the same traffic participant is sorted according to the reporting time, so as to obtain the traffic track information of the traffic participant. 3. The method according to claim 1, wherein determining the adjacency relationship and the number of passes of any two adjacent traffic points in the traffic trajectory according to the traffic trajectory information, comprising: splitting the traffic trajectory information into adjacent traffic point pairs; Determine the adjacency relationship of any two adjacent traffic points according to the adjacent traffic point pair; Determine the number of track passes of any two adjacent traffic points according to the adjacent traffic point pairs. 4. The method according to claim 1, characterized in that, based on the information of the traffic point and the adjacency relationship between any two adjacent traffic points and the number of track passes, the method obtained by the spectral clustering method is obtained by the said traffic point. A high-dimensional representation of the traffic network composed of traffic points, including: The adjacency matrix is determined by taking the number of track passes of any two adjacent traffic points as the weight value of the adjacency relationship of the corresponding two adjacent traffic points; According to the information of the traffic point, the adjacency relationship between any two adjacent traffic points and the adjacency matrix, obtain a weighted undirected adjacency graph of the traffic network composed of the traffic points; From the weighted undirected adjacency graph, a high-dimensional representation of the traffic network consisting of the traffic points is computed. 5 . The method according to claim 4 , wherein the calculating a high-dimensional representation of the traffic network composed of the traffic points according to the weighted undirected adjacency graph, comprising: 6 . According to the weighted undirected adjacency graph, the degrees corresponding to all traffic points are calculated to obtain a degree matrix; based on the degree matrix and the adjacency matrix, calculating a Laplace matrix of the weighted undirected adjacency graph; solve the eigenvectors and eigenvalues of the Laplacian matrix; According to the eigenvectors and eigenvalues of the Laplacian matrix, using the TopK algorithm, a high-dimensional representation of the traffic network composed of the traffic points is obtained. 6. The method according to claim 1, wherein the method further comprises: Receive requests for information on specific categories of traffic points; Based on the category information in the request, the information of the traffic point of the corresponding category is obtained from the database to respond. 7. A device for processing traffic information, wherein the device comprises: The acquisition unit is used to acquire the information of the traffic point; a conversion unit, configured to convert the information of the traffic point into traffic track information; a determining unit, configured to determine the adjacency relationship between any two adjacent traffic points in the traffic track and the number of track passes according to the traffic track information; a computing unit, configured to obtain a high-dimensional representation of the traffic network composed of the traffic points by using the method of spectral clustering based on the information of the traffic points and the adjacency relationship between any two adjacent traffic points and the number of passes; an extraction unit, configured to extract the category information of the traffic point from the high-dimensional representation; The storage unit is configured to classify and store the information of the traffic point based on the category information to form a database. 8. The apparatus according to claim 7, wherein the computing unit comprises: The adjacency matrix calculation module is used to determine the adjacency matrix by taking the number of track passes of any two adjacent traffic points as the weight value of the adjacency relationship between the corresponding two adjacent traffic points; an adjacency graph calculation module, configured to obtain a weighted undirected adjacency graph of the traffic network composed of the traffic points according to the information of the traffic points, the adjacency relationship between any two adjacent traffic points, and the adjacency matrix; A high-dimensional representation calculation module, configured to calculate a high-dimensional representation of the traffic network composed of the traffic points according to the weighted undirected adjacency graph. 9. A terminal, comprising a memory, a processor and a computer program stored in the memory and running on the processor, wherein the processor implements the computer program as claimed in claim 1 when the processor executes the computer program The steps of any one of to 6. 10. A computer-readable storage medium storing a computer program, characterized in that, when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 6 are implemented .

Images