Graph segmentation algorithm-based crowdsourcing map forming method and system, medium and equipment

By using graph segmentation algorithm in crowdsourcing map formation, the visual odometer and loop detection data are processed, and segmented sub-graphs are generated in parallel, the information loss and calculation pressure problems in crowdsourcing map formation are solved, and global consistency is achieved Map generation.

CN119027661APending Publication Date: 2024-11-26WUHAN ZHONGHAITING DATA TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410918173.5
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-07-10
Publication Date
2024-11-26

AI Technical Summary

Technical Problem

In crowdsourcing map mapping scenarios, it is difficult for the existing technology to establish globally consistent electronic maps while reducing information losses and alleviating computing pressure, especially under the requirements of large-scale data acquisition and real-time updates, computing resources and The update frequency issue has not been fully considered.

Method used

Using a graph segmentation algorithm, by obtaining the visual odometer measurement data after calibration of GPS signal, loopback detection and constructing a rights-free undirected graph, using the LDG algorithm to segment the graph, generate multiple segmented sub-graphs, and then Enter the graph model for parallel processing to generate an electronic map.

Benefits of technology

On the premise of minimizing information losses as much as possible and alleviating computing pressure, electronic map generation that maintains global consistency within a certain collection range is achieved, improving the computing efficiency and feasibility of data processing.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119027661A_ABST
    Figure CN119027661A_ABST
Patent Text Reader

Abstract

The invention discloses a crowdsourcing map forming method and system based on a graph segmentation algorithm, a medium and equipment. The method comprises the following steps: acquiring visual odometer measurement data after GPS signal calibration of a road to be processed; obtaining a loopback matching result of each driving track in the measurement data of the visual odometer based on a loopback detection algorithm; constructing a weighted undirected graph according to the loopback matching result; segmenting the weighted undirected graph based on an LDG algorithm to obtain a plurality of segmented sub-graphs; generating an electronic map of the road to be processed according to the plurality of segmentation sub-graphs; therefore, on the premise that information loss is reduced as much as possible and calculation pressure is relieved, the electronic map which keeps global consistency within a certain collection range can be established.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of high-precision map technology, and in particular to a crowdsourcing map generation method, system, medium and device based on a graph segmentation algorithm. Background Art

[0002] Autonomous driving technology has extremely high commercial application value. After years of development, many different technical routes and solutions have emerged. The same goal is to enable cars to realize the intelligent "perception-decision-execution" process. Among them, high-precision electronic maps are an important basis for cars to perceive and make decisions. This field is highly related to the traditional simultaneous localization and mapping problem. The SLAM mapping process can be divided into several steps, including visual odometer, global optimization, loop detection and mapping.

[0003] According to the data source used by the visual odometer, the mapping route can be divided into two types: special package and crowdsourcing. Special package data appeared in the early stage of the development of traditional navigation, which requires a large number of professional data collection personnel and highly sophisticated mapping equipment. Due to various cost reasons, it cannot well meet the needs of large-scale mapping and timely updates required by autonomous driving. Therefore, the current mainstream solution tends to choose a data source based on crowdsourcing data.

[0004] Crowdsourcing data can be understood as users obtaining motion information through ordinary sensors and other low-cost hardware on their cars. After a large number of cars repeatedly collect information on a section of road, relatively accurate map data is obtained through image recognition and statistical knowledge or deep learning solutions in the field of AI. In principle, it is to make up for the problem of low quality and accuracy of individual data through large-scale data sharing, mining, analysis and fusion, and improve map accuracy and credibility. Its advantage is low cost, and ordinary vehicles can perform mapping tasks after simple modification. The data source is rich and the real-time performance is good.

[0005] In crowdsourcing scenarios, the map data usually required is massive and updated in real time, which places high demands on the processing capabilities of backend optimization. Currently, the mainstream crowdsourcing mapping solutions in academia rarely consider issues such as computing resources and update frequency, which are actually crucial for commercialization. Summary of the invention

[0006] The present invention provides a crowdsourcing map generation method, system, medium and equipment based on a graph segmentation algorithm, which can establish an electronic map that maintains global consistency within a certain collection range under the premise of minimizing information loss and alleviating computing pressure.

[0007] In a first aspect, a crowdsourcing map generation method based on a graph segmentation algorithm is provided, comprising the following steps:

[0008] Obtain visual odometer measurement data after GPS signal calibration of the road to be processed;

[0009] Obtaining loop matching results of each driving trajectory in the visual odometer measurement data based on a loop detection algorithm;

[0010] Constructing a weighted undirected graph according to the loop matching result;

[0011] Based on the LDG algorithm, the weighted undirected graph is segmented to obtain a plurality of segmented subgraphs;

[0012] An electronic map of the road to be processed is generated according to the plurality of segmented sub-maps.

[0013] According to the first aspect, in a first possible implementation manner of the first aspect, the step of “constructing a weighted undirected graph according to the loop matching result” specifically includes the following steps:

[0014] A weighted undirected graph is constructed with a single driving trajectory as a vertex, the matching relationships in the loop matching results of each driving trajectory as edges, and the matching number in the loop matching results of each driving trajectory as the edge weight.

[0015] According to the first aspect, in a second possible implementation manner of the first aspect, the LDG algorithm index(v) is:

[0016]

[0017] Among them, the scoring function w(i,t) is:

[0018]

[0019] Where P i is the i-th segmentation subgraph; N(v) is the neighbors of the current vertex; C is the preset maximum load; |P i | is the degree of the i-th segmentation subgraph; K is the number of segmentation subgraphs.

[0020] According to the first aspect, in a third possible implementation manner of the first aspect, the step of “generating an electronic map of the road to be processed according to the plurality of segmented sub-graphs” specifically includes the following steps:

[0021] The visual odometer measurement data, loop matching results and vehicle driving GNSS data of the plurality of segmented sub-graphs are input into a mapping model, and each segmented sub-graph is mapped separately in parallel based on a mapping algorithm to obtain an electronic map of the road to be processed.

[0022] According to the third possible implementation manner of the first aspect, in a fourth possible implementation manner of the first aspect, the mapping algorithm is:

[0023]

[0024] In the formula, S i , S j are the positions of different trajectory points respectively; r o is the visual odometry residual factor; Measure data for visual odometry; GNSS measurement for vehicle driving; is the loop matching result; r g is the GNSS residual factor; G is the GNSS signal area; r l is the loop residual factor; L is the area where the loop exists.

[0025] In the second aspect, a crowdsourcing map generation system based on a graph segmentation algorithm is provided, comprising:

[0026] The odometer module is used to obtain the visual odometer measurement data after the GPS signal calibration of the road to be processed;

[0027] A loop module, which is in communication with the odometer module and is used to obtain a loop matching result of each driving track in the visual odometer measurement data based on a loop detection algorithm;

[0028] A weighted undirected graph module, which is in communication with the loop module and is used to construct a weighted undirected graph according to the loop matching result;

[0029] a segmentation module, which is in communication connection with the weighted undirected graph module and is used to segment the weighted undirected graph based on the LDG algorithm to obtain a plurality of segmented subgraphs; and

[0030] The mapping module is in communication with the segmentation module and is used to generate an electronic map of the road to be processed according to the plurality of segmentation sub-maps.

[0031] According to a third aspect, a computer-readable storage medium is provided, on which a computer program is stored, wherein when the computer program is executed by a processor, the crowdsourcing map generation method based on the graph segmentation algorithm as described above is implemented.

[0032] In a fourth aspect, an electronic device is provided, comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, wherein the processor implements the crowdsourcing map generating method based on the graph segmentation algorithm as described above when running the computer program.

[0033] Compared with the prior art, the advantages of the present invention are as follows: a weighted undirected graph is segmented to obtain multiple segmented sub-graphs; and an electronic map of the road to be processed is generated based on the multiple segmented sub-graphs. Therefore, a feasible solution for crowdsourcing data back-end mapping is given based on the graph segmentation algorithm in the field of graph theory - the LDG algorithm, which can establish an electronic map that maintains global consistency within a certain collection range under the premise of minimizing information loss and alleviating computing pressure. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 It is a flow chart of an embodiment of a crowdsourcing map generation method based on a graph segmentation algorithm of the present invention;

[0035] Figure 2 It is a schematic diagram of the authorized undirected graph of the present invention;

[0036] Figure 3 It is a schematic diagram of segmentation of a weighted undirected graph of the present invention;

[0037] Figure 4 It is a structural schematic diagram of a crowdsourcing map generation system based on a graph segmentation algorithm of the present invention. DETAILED DESCRIPTION

[0038] Reference will now be made in detail to specific embodiments of the present invention, examples of which are illustrated in the accompanying drawings. Although the present invention will be described in conjunction with specific embodiments, it will be understood that it is not intended to limit the present invention to the embodiments described. On the contrary, it is intended to cover changes, modifications and equivalents included in the spirit and scope of the present invention as defined by the appended claims. It should be noted that the method steps described herein can all be implemented by any functional block or functional arrangement, and any functional block or functional arrangement can be implemented as a physical entity or a logical entity, or a combination of the two.

[0039] In order to enable those skilled in the art to better understand the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific implementation methods.

[0040] Note: The example to be introduced below is only a specific example, and is not intended to limit the embodiments of the present invention to the following specific steps, values, conditions, data, sequences, etc. Those skilled in the art can apply the concept of the present invention to construct more embodiments not mentioned in this specification by reading this specification.

[0041] See also Figure 1 As shown, an embodiment of the present invention provides a crowdsourcing map generation method based on a graph segmentation algorithm, comprising the following steps:

[0042] S100, obtaining visual odometer measurement data after GPS signal calibration of the road to be processed;

[0043] Visual odometry deals with the problem of camera motion between adjacent images. It can use the video data obtained from the on-board camera to estimate the camera motion through the image relationship between adjacent frames, and further restore the spatial structure of the scene. Commonly used methods include direct method, optical flow method, feature point method, etc. It is called an odometer because it only calculates the motion at adjacent moments and has no connection with the previous information. By connecting the motion between each two frames in series, the motion trajectory of the vehicle with the camera installed is obtained. Since the error of the odometer will accumulate, global data from another source is needed to eliminate this part of the error. A common tool on a moving car is the GPS signal. The odometer result with the addition of the GPS signal can reduce the cumulative error.

[0044] S200, obtaining a loop matching result of each driving track in the visual odometer measurement data based on a loop detection algorithm;

[0045] Loop detection also solves the problem of position estimation drifting over time. In the scenario of autonomous driving, it provides the judgment that "the car has passed the same position at two time points". The specific judgment method is mainly based on the similarity between the pictures taken by the on-board camera. For example, if a car passes the same intersection twice, the traffic lights at the intersection, the turn signs and stop lines on the ground, nearby buildings and guardrails can all be identified as a semantic feature in the image. If a large number of overlapping semantic features appear in the two pictures of the dashcam, it can be judged that they are most likely taken at the same location. For example, car a passes an intersection at time t1, and the semantic features are used to judge that car b passes the same intersection at time t2, and then it is obtained that car a at time t1 and car b at time t2 are at the same position. That is, a loop is determined between car a at time t1 and car b at time t2, or it is called a match between car a and car b.

[0046] S300, constructing a weighted undirected graph according to the loop matching result;

[0047] S400, segmenting the weighted undirected graph based on the LDG algorithm to obtain a plurality of segmented subgraphs;

[0048] S500: Generate an electronic map of the road to be processed according to the plurality of segmented sub-maps.

[0049] Specifically, in this embodiment, the present invention divides the weighted undirected graph into multiple divided sub-graphs; then generates an electronic map of the road to be processed based on the multiple divided sub-graphs. Therefore, a graph segmentation algorithm based on the graph theory field - LDG (Linear Deterministic Greedy partitioning) algorithm is used to provide a feasible solution for crowdsourcing data back-end mapping, which can establish an electronic map that maintains global consistency within a certain collection range under the premise of minimizing information loss and alleviating computing pressure.

[0050] Preferably, in another embodiment of the present application, the step of "S300, constructing a weighted undirected graph according to the loop matching result" specifically includes the following steps:

[0051] A weighted undirected graph is constructed with a single driving trajectory as the vertex, the matching relationship in the loop matching results of each driving trajectory as the edge, and the matching number in the loop matching results of each driving trajectory as the edge weight.

[0052] Specifically, in this embodiment, in the crowdsourcing scenario, there are a large number of trajectories, and there are often matches between two trajectories. If each trajectory is a vertex, an edge is connected between two trajectories with a matching relationship, and the edge weight is the number of matching relationships between the two trajectories, a weighted undirected graph G can be established, see Figure 2 As shown in the figure, graph G represents the structure of the trajectory problem to be solved. If a batch of collected data is basically in the same area, there is a high probability that graph G contains only one connected component, which means that any two trajectories of this batch of data can be connected by matching with at least one other trajectory.

[0053] Figure 2 The typical structure of a graph G is shown, which contains five trajectories AE, one loop between AB trajectories, and four loops between CE. So they are connected to each other to form a connected component.

[0054] If this batch of data is added to the backend optimization, the required computing resources will increase linearly with the increase in the collection scale, but the memory and server scale cannot increase linearly during the actual map construction process, so it is necessary to appropriately split the graph G. Because the results after segmentation are directly entered into the backend optimization, it is expected that the scale of each subgraph after segmentation is roughly the same, and the matching edges between subgraphs are as few as possible.

[0055] Preferably, in another embodiment of the present application, the LDG algorithm index(v) is:

[0056]

[0057] Among them, the scoring function w(i,t) is:

[0058]

[0059] Where P i is the i-th segmentation subgraph; N(v) is the neighbors of the current vertex; C is the preset maximum load; |P i | is the degree of the i-th segmentation subgraph; K is the number of segmentation subgraphs.

[0060] Specifically, in this embodiment, in graph theory, the problem of satisfying both the load balancing condition (the structure of the divided subgraphs is uniform, and there will not be a subgraph with only one vertex and another subgraph with 100 vertices) and the minimum cut edge (the sum of the weights of the edges discarded by the division is as small as possible) is called the graph partitioning problem, which is an NP-complete problem. That is, there is no better solution to this problem except the exponential time complexity. The advantage of the LDG algorithm is that its result does not depend on the specific traversal order, and it has better time performance and cut set loss marking for large tasks. The specific method is to traverse the entire graph from an initial node, and perform the following operations on each vertex visited for the first time, score the proximity of the vertex to all current core vertices, take the core vertex with the highest score, and add the vertex to be partitioned to this core vertex. For the specific algorithm formula, see index(v).

[0061] Take Figure G as an example. Figure 3 As shown in the figure, after LDG partitioning (the dividing line in the figure), the graph is divided into two parts, the partitioned subgraph ACE and the partitioned subgraph BD. In this way, ACE can be used as the first subproblem, in which only the matching relationship between ACEs needs to be considered. BD is similar as the second subproblem. Therefore, the problem can be divided into two smaller subproblems for solution while retaining the matching relationship as much as possible and the subgraph size is uniform. For large-scale problems, there may be hundreds of nodes. This division method can effectively retain the original data and reduce the scale of the problem. At the same time, since the subgraphs are of similar size, it is conducive to parallel optimization. Parallelism means that these subproblems do not interfere with each other and can be run simultaneously on multiple computers or multiple processors on a computer. This improves the solution efficiency.

[0062] Preferably, in another embodiment of the present application, the step of "S500, generating an electronic map of the road to be processed according to the plurality of segmented sub-graphs" specifically includes the following steps:

[0063] The visual odometer measurement data, loop matching results and vehicle driving GNSS data of the plurality of segmented sub-graphs are input into a mapping model, and each segmented sub-graph is mapped separately in parallel based on a mapping algorithm to obtain an electronic map of the road to be processed.

[0064] Specifically, in this embodiment, according to the process of the SLAM mapping algorithm, the observed odometer data, GNSS data and loop closure data are input into the mapping model to obtain the mapping result. A typical mapping algorithm is pose graph optimization. The pose of each trajectory point is used as a node, and the GNSS constraint, odometer constraint and loop closure constraint are used as edges. Find the trajectory pose point that satisfies the following minimum value:

[0065]

[0066] In the formula, Si , S j are the positions of different trajectory points respectively; r o is the visual odometry residual factor; Measure data for visual odometry; GNSS measurement for vehicle driving; is the loop matching result; r g is the GNSS residual factor; G is the GNSS signal area; r l is the loop residual factor; L is the area where the loop exists.

[0067] See also Figure 4 As shown, an embodiment of the present invention provides a crowdsourcing map generation system based on a graph segmentation algorithm, comprising:

[0068] The odometer module is used to obtain the visual odometer measurement data after the GPS signal calibration of the road to be processed;

[0069] A loop module, which is in communication with the odometer module and is used to obtain a loop matching result of each driving track in the visual odometer measurement data based on a loop detection algorithm;

[0070] A weighted undirected graph module, which is in communication with the loop module and is used to construct a weighted undirected graph according to the loop matching result;

[0071] a segmentation module, which is in communication connection with the weighted undirected graph module and is used to segment the weighted undirected graph based on the LDG algorithm to obtain a plurality of segmented subgraphs; and

[0072] The mapping module is in communication with the segmentation module and is used to generate an electronic map of the road to be processed according to the plurality of segmentation sub-maps.

[0073] Therefore, the present invention divides the weighted undirected graph into multiple divided sub-graphs; and then generates an electronic map of the road to be processed based on the multiple divided sub-graphs. Therefore, based on the graph segmentation algorithm in the field of graph theory - LDG algorithm, a feasible solution for crowdsourcing data back-end mapping is given, which can establish an electronic map that maintains global consistency within a certain collection range under the premise of minimizing information loss and alleviating computing pressure.

[0074] Specifically, this embodiment corresponds one-to-one to the above method embodiment, and the functions of each module have been described in detail in the corresponding method embodiment, so they will not be repeated here.

[0075] Based on the same inventive concept, an embodiment of the present application further provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, all or part of the method steps of the above method are implemented.

[0076] The present invention implements all or part of the process in the above method, and can also be completed by instructing related hardware through a computer program. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, the steps of each method embodiment can be implemented. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, USB flash drive, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can 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 medium does not include electric carrier signals and telecommunication signals.

[0077] Based on the same inventive concept, an embodiment of the present application also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program running on the processor, and when the processor executes the computer program, all or part of the method steps in the above method are implemented.

[0078] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (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 any conventional processor, etc. The processor is the control center of a computer device, and uses various interfaces and lines to connect various parts of the entire computer device.

[0079] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the computer device by running or executing the computer programs and / or modules stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system and at least one application required for a function (such as a sound playback function, an image playback function, etc.); the data storage area can store data created according to the use of the mobile phone (such as audio data, video data, etc.). In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0080] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, systems, servers or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program codes.

[0081] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), servers, and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as a combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0082] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0083] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process. Figure 1 A process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0084] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A crowdsourcing map generation method based on a graph segmentation algorithm, characterized in that: The following steps are involved: Obtain visual odometer measurement data after GPS signal calibration of the road to be processed; Obtaining loop matching results of each driving trajectory in the visual odometer measurement data based on a loop detection algorithm; Constructing a weighted undirected graph according to the loop matching result; Based on the LDG algorithm, the weighted undirected graph is segmented to obtain a plurality of segmented subgraphs; An electronic map of the road to be processed is generated according to the plurality of segmented sub-maps.

2. The crowdsourcing map generation method based on the graph segmentation algorithm according to claim 1, characterized in that: The step of "constructing a weighted undirected graph according to the loop matching result" specifically includes the following steps: A weighted undirected graph is constructed with a single driving trajectory as a vertex, the matching relationships in the loop matching results of each driving trajectory as edges, and the matching number in the loop matching results of each driving trajectory as the edge weight.

3. The crowdsourcing map generation method based on the graph segmentation algorithm according to claim 1, characterized in that: The LDG algorithm index(v) is: Among them, the scoring function w(i,t) is: Where P i is the i-th segmentation subgraph; N(v) is the neighbors of the current vertex; C is the preset maximum load; |P i | is the degree of the i-th segmentation subgraph; K is the number of segmentation subgraphs.

4. The crowdsourcing map generation method based on the graph segmentation algorithm according to claim 1, characterized in that: The step of "generating an electronic map of the road to be processed according to the plurality of segmented sub-graphs" specifically comprises the following steps: The visual odometer measurement data, loop matching results and vehicle driving GNSS data of the plurality of segmented sub-graphs are input into a mapping model, and each segmented sub-graph is mapped separately in parallel based on a mapping algorithm to obtain an electronic map of the road to be processed.

5. The crowdsourcing map generation method based on the graph segmentation algorithm according to claim 4 is characterized in that: The mapping algorithm is: In the formula, S i , S j are the positions of different trajectory points respectively; r o is the visual odometry residual factor; Measure data for visual odometry; GNSS measurement for vehicle driving; is the loop matching result; r g is the GNSS residual factor; G is the GNSS signal area; r l is the loop residual factor; L is the area where the loop exists.

6. A crowdsourcing map generation system based on a graph segmentation algorithm, characterized in that: include: The odometer module is used to obtain the visual odometer measurement data after the GPS signal calibration of the road to be processed; A loop module, which is in communication with the odometer module and is used to obtain a loop matching result of each driving track in the visual odometer measurement data based on a loop detection algorithm; A weighted undirected graph module, which is in communication with the loop module and is used to construct a weighted undirected graph according to the loop matching result; A segmentation module, which is in communication connection with the weighted undirected graph module and is used to segment the weighted undirected graph based on the LDG algorithm to obtain a plurality of segmented subgraphs; as well as, The mapping module is in communication with the segmentation module and is used to generate an electronic map of the road to be processed according to the plurality of segmentation sub-maps.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the crowdsourcing map generation method based on a graph segmentation algorithm as described in any one of claims 1 to 5 is implemented.

8. An electronic device comprising a storage medium, a processor, and a computer program stored in the storage medium and executable on the processor, characterized in that: When the processor runs the computer program, it implements the crowdsourcing map generation method based on the graph segmentation algorithm as described in any one of claims 1 to 5.