Efficient multi-packet matching fusion method and device based on connected graph, medium and equipment

Abstract

The application discloses a multi-pack efficient matching fusion method based on a connected graph, and belongs to the field of map data processing. The method comprises the following steps: in K low-precision single-pack maps passing through a same center point, two subgraphs corresponding to a same center point in each two adjacent low-precision single-pack maps are automatically matched, and the same center point in the K low-precision single-pack maps is sequentially connected, so that a matching score corresponding to different data elements between each two subgraphs in a complete connected graph is obtained, and an object corresponding to a most confident matching meeting geometric consistency is determined according to the matching score; and the two subgraphs passing through a same center point in each low-precision single-pack map and another low-precision single-pack map with the largest overlapping area are automatically matched, so that an object corresponding to a most confident matching meeting geometric consistency is determined. The application saves the equipment cost, reduces the complexity of calculation, guarantees the stability of matching, and does not depend on the uniqueness of map elements.

Description

Technical Field

[0001] This application relates to the field of map data processing, and in particular to a method, apparatus, medium and device for efficient multi-packet matching and fusion based on connected graphs. Background Technology

[0002] In existing technologies, rule-based methods rely heavily on the uniqueness of map elements when processing map data. These methods first identify the most unique objects within a single map packet, then match these unique objects together, and so on, finding and matching less unique objects. Matching failures can occur if the single map packet lacks unique elements, or if unique elements are missing (e.g., in tunnel scenes or scenes without signs or poles); if unique elements are poorly positioned (e.g., improperly placed signs or poles); or if a large amount of map data is discarded during the matching process. Rule-based methods also rely on thresholds, requiring continuous adjustment for different scenarios, such as determining the range within which an object is considered unique.

[0003] This solution utilizes low-precision equipment to generate high-precision maps through learning and computation, thus solving the problems of high cost of high-precision equipment and reliance on the uniqueness of map elements in existing technologies. Summary of the Invention

[0004] To address the problems of excessive reliance on the uniqueness of map elements and high cost of obtaining high-precision maps in existing technologies, this application mainly provides a method, apparatus, medium, and device for efficient multi-packet matching and fusion based on connected graphs.

[0005] One technical solution adopted in this application is: providing a multi-packet efficient matching and fusion method based on connected graphs, which includes:

[0006] Get N low-precision single-packet maps of the same task that do not have overlapping data elements, where N is a natural number greater than 1;

[0007] In each low-precision single-package map, a sub-map with a fixed range is generated with a randomly selected data element as the center point;

[0008] In K low-precision single-packet maps that pass through the same center point, the two subgraphs corresponding to the same center point in each of the two adjacent low-precision single-packet maps are automatically matched, and the same center point in the K low-precision single-packet maps is connected in sequence to obtain the matching score corresponding to different data elements between each two subgraphs in the complete connected graph. The object corresponding to the most certain matching that meets the geometric consistency is determined by the matching score, where K is a natural number less than N and greater than 0.

[0009] Each low-precision single-packet map is automatically matched with the two sub-maps that pass through the same center point in the other low-precision single-packet map with the largest overlapping area. The matching scores are obtained for different data elements between the two sub-maps. The objects corresponding to the most certain matching that meet geometric consistency are determined by the matching scores.

[0010] Based on the most certain match, obtain the objects corresponding to other most certain matches that are geometrically consistent and are adjacent to the most certain match;

[0011] Align the objects that are most confidently matched between each pair of low-precision single-packet maps one by one to obtain a high-precision single-packet map.

[0012] By merging and optimizing high-precision single-package maps, a high-precision map is obtained.

[0013] Another technical solution adopted in this application is: providing a multi-packet high-efficiency matching and fusion device based on a connected graph, which includes:

[0014] A module for obtaining a low-precision single-packet map of N non-overlapping data elements in the same task, where N is a natural number greater than 1;

[0015] A module used to generate a sub-map with a fixed range in each low-precision single-package map, with randomly selected data elements as the center point;

[0016] This is used to automatically match the two subgraphs corresponding to the same center point in each of the K low-precision single-packet maps that pass through the same center point, and then connect the same center point in the K low-precision single-packet maps in sequence to obtain the matching score corresponding to different data elements between each two subgraphs in the complete connected graph. The module corresponding to the object with the most certain matching that meets geometric consistency is determined by the matching score, where K is a natural number less than N and greater than 0.

[0017] This module is used to automatically match the two sub-maps that each low-precision single-packet map passes through with the same center point in the other low-precision single-packet map with the largest overlapping area, obtain the matching score corresponding to different data elements between the two sub-maps, and determine the object corresponding to the most certain match that meets geometric consistency through the matching score.

[0018] A module for obtaining objects that are geometrically consistent and adjacent to the most certain match, based on the most certain match;

[0019] This module is used to align the objects that are most confidently matched between any two low-precision single-packet maps one by one to obtain a high-precision single-packet map.

[0020] Another technical solution adopted in this application is to provide a computer-readable storage medium storing computer instructions that are operated to execute the multi-packet efficient matching and fusion method based on connected graphs in Solution 1.

[0021] Another technical solution adopted in this application is: providing a computer device, which includes a processor and a memory, the memory storing computer instructions, which are operated to execute the multi-packet efficient matching and fusion method based on connected graphs in Solution 1.

[0022] The beneficial effects of the technical solution of this application are as follows: This application designs a method, apparatus, medium, and device for efficient multi-packet matching and fusion based on connected graphs. This application saves equipment costs by fusing low-precision single-packet maps into high-precision maps through matching, reduces computational complexity by matching K low-precision single-packet maps, ensures the stability of matching, and is most confident that the matching does not depend on the uniqueness of map elements. Attached Figure Description

[0023] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0024] Figure 1 This is a schematic diagram of a specific implementation of a multi-packet efficient matching and fusion method based on connected graphs according to this application;

[0025] Figure 2 This is a schematic diagram of a specific embodiment of a multi-packet efficient matching and fusion device based on a connected graph according to this application.

[0026] The accompanying drawings have illustrated specific embodiments of this application, which will be described in more detail below. These drawings and descriptions are not intended to limit the scope of the concept in any way, but rather to illustrate the concept of this application to those skilled in the art through reference to specific embodiments. Detailed Implementation

[0027] The preferred embodiments of this application will now be described in detail with reference to the accompanying drawings, so that the advantages and features of this application can be more easily understood by those skilled in the art, thereby providing a clearer and more definite definition of the scope of protection of this application.

[0028] It should be noted that, in this document, relational terms such as "first" and "second" are used merely to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising..." does not exclude the presence of additional identical elements in the process, method, article, or apparatus that includes said element.

[0029] The technical solution of this application and how the technical solution of this application solves the above-mentioned technical problems are described in detail below with specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments. The embodiments of this application will be described below with reference to the accompanying drawings.

[0030] Figure 1 This paper illustrates a specific implementation of a multi-packet efficient matching and fusion method based on connected graphs, as proposed in this application. Figure 1 In the specific implementation shown, the efficient multi-packet matching and fusion method based on connected graphs includes:

[0031] Step S101: Obtain N low-precision single-packet maps of the same task that do not contain overlapping data elements, where N is a natural number greater than 1.

[0032] In this implementation, a task can create multiple low-precision single-package maps. However, in order to prevent the same path from being created twice, the low-precision single-package map containing backtracking needs to be split at the backtracking point to create two low-precision single-package maps that do not contain backtracking, which facilitates subsequent matching.

[0033] It should be noted that a low-precision single-package map refers to a vector map that collects spatial and semantic information about traffic elements such as traffic signs and lane lines by a vehicle traveling along a section of road.

[0034] In a specific example of this application, a low-precision device can collect data in one pass to build a low-precision single-package map. All data collected in one pass belongs to the information within this low-precision single-package map. The low-precision single-package maps built from multiple passes are then matched, fused, and optimized. Because low-precision data collection devices, such as GPS, have low accuracy and limited computing power, the perception results are unreliable. This method uses a pairwise pairing approach; the same object is paired with only one type of data element in the low-precision single-package map. If a low-precision single-package map has a backtracking mechanism, it's equivalent to traversing the same location twice, building maps twice, and having two data elements. The same object will correspond to two data elements represented in the low-precision single-package map.

[0035] exist Figure 1 In the specific implementation shown, the multi-packet efficient matching and fusion method based on connected graphs further includes:

[0036] Step S102: In each low-precision single-package map, a sub-map with a fixed range is generated, centered on a randomly selected data element.

[0037] In this embodiment, a low-precision single-packet map contains many sub-maps. The low-precision single-packet map is of variable size, while the sub-maps are of fixed size. The purpose of the sub-maps is to facilitate matching. There is a large amount of overlap between multiple sub-maps, which is intended to provide redundant matching in the future and ensure the robustness of the matching.

[0038] In one specific embodiment of this application, obtaining N low-precision single-packet maps that do not contain overlapping data elements in the same task includes: if the low-precision single-packet map contains backtracking paths, then splitting it at the backtracking paths to obtain two low-precision single-packet maps that do not contain overlapping data elements.

[0039] In this embodiment, the same task contains multiple low-precision single-package maps. Some low-precision single-package maps contain backtracking paths, while others do not. In order to facilitate the matching of map elements, the low-precision single-package maps containing backtracking paths need to be segmented at the backtracking paths so that each object corresponds to one map element in a low-precision single-package map.

[0040] exist Figure 1 In the specific implementation shown, the multi-packet efficient matching and fusion method based on connected graphs further includes:

[0041] Step S103: In the K low-precision single-packet maps that pass through the same center point, the two subgraphs corresponding to the same center point in each of the two adjacent low-precision single-packet maps are automatically matched, and the same center point in the K low-precision single-packet maps is connected in sequence to obtain the matching score corresponding to different data elements between each two subgraphs in the complete connected graph. The object corresponding to the most certain match that meets the geometric consistency is determined by the matching score, where K is a natural number less than N and greater than 0.

[0042] In this embodiment, every two adjacent pairs in the K are matched to obtain a complete connected graph, thereby reducing the computational complexity.

[0043] In a specific instance of this application, a task has N low-precision single-packet maps. Pairing them one by one would result in a complexity of N^2, so a K+1 matching method is adopted. There are multiple low-precision single-packet maps passing through the center point of the same object; 1-2-3-4 can be understood as four low-precision single-packet maps each. For the same center point, forming a connected graph requires that all low-precision single-packet maps passing through the same center point must be together; there cannot be a missing low-precision single-packet map. K low-precision single-packet maps must satisfy the condition of a connected graph; the selected K low-precision single-packet maps must be connected based on each center point. When low-precision single-packet maps... Figure 2 If the match with low-precision single-pack map 3 is not good, it will lead to low-precision single-pack map 4 and low-precision single-pack map 5 being poorly matched. Figure 2 There will also be deviations.

[0044] In one specific embodiment of this application, determining the object corresponding to the most certain match that meets geometric consistency by matching score includes: when the matching score is consistent with a benchmark threshold, the data element corresponding to the matching score meets geometric consistency; and when the proportion of the number of matching scores that are consistent with the benchmark threshold in the matching score is greater than a preset threshold, the data element corresponding to the matching score is the object corresponding to the most certain match.

[0045] In this embodiment, multiple matching results can be filtered out using a benchmark threshold. All multiple matching results meet the geometric consistency requirement. When the proportion of the benchmark threshold's number to the total number of matching scores reaches a preset threshold, it is considered the most certain match. This ensures the accuracy of the matching.

[0046] exist Figure 1 In the specific implementation shown, the multi-packet efficient matching and fusion method based on connected graphs further includes:

[0047] Step S104: Automatically match the two sub-maps that each low-precision single-packet map passes through with the same center point in the other low-precision single-packet map with the largest overlapping area, and obtain the matching score corresponding to different data elements between the two sub-maps. The object corresponding to the most certain match that meets the geometric consistency is determined by the matching score.

[0048] In this embodiment, each low-precision single-packet map is re-matched with the other low-precision single-packet map with the largest overlapping area to ensure the stability of the matching.

[0049] In one specific embodiment of this application, automatic matching is performed on two sub-maps traversed by the same center point in each low-precision single-packet map and the other low-precision single-packet map with the largest overlap area. This includes: selecting the two sub-maps with the largest overlap area in the two low-precision single-packet maps based on the number of sub-maps traversed by the same center point in each of the two low-precision single-packet maps; simultaneously extracting features from different data elements between the two sub-maps with the largest overlap area to obtain feature information corresponding to the two sub-maps; and inputting the feature information corresponding to the two sub-maps into an attention map neural network for automatic matching to obtain matching scores corresponding to different data elements between the two sub-maps.

[0050] In this embodiment, the information of the data elements to be matched is extracted to lay the foundation for subsequent matching; different data elements in each sub-graph can be extracted simultaneously. When matching, the feature information of the data elements in the corresponding sub-graphs of the same center point in two low-precision single-package maps is matched simultaneously, which is efficient and convenient.

[0051] In one specific embodiment of this application, feature extraction is performed simultaneously on different data elements between the two subgraphs with the largest overlapping area to obtain their respective feature information. This includes: extracting features from different data elements in the same subgraph based on the differences in the surrounding elements and their own attributes. When a data element belongs to the first category, features are extracted from its surrounding elements and its own attributes to obtain corresponding semantic feature information and perceptual feature information. The semantic feature information and perceptual feature information are then superimposed to obtain the overall feature information corresponding to the data element. When another data element belongs to the second category, features are extracted from its own attributes to obtain attribute feature information. This process yields feature information corresponding to different types of data elements in the same subgraph.

[0052] In this embodiment, feature extraction is performed on different data elements in each subgraph. Feature extraction makes matching data elements easier and more convenient. The semantic feature information in the feature information includes, but is not limited to, the type, size, relative depth, and orientation of other elements surrounding the data element; the perceptual feature information and attribute feature information in the feature information include the position coordinates of the data element.

[0053] In one specific embodiment of this application, the feature information corresponding to two sub-graphs is input into an attention graph neural network for automatic matching to obtain matching scores corresponding to different data elements between the two sub-graphs. This includes: the attention graph neural network automatically learns attention on the feature information of the data elements corresponding to the two sub-graphs, and simultaneously automatically learns attention on the feature information between different data elements in the same sub-graph to obtain matching scores corresponding to different data elements between the two sub-graphs.

[0054] In this embodiment, the attention graph neural network is pre-trained and has high computational efficiency. Different attention scores are assigned based on the attention mechanism to identify more important correspondences. The attention graph neural network automatically learns the attention between different types of data elements in the same subgraph, enabling it to better identify objects corresponding to different data elements in the same subgraph; the attention graph neural network also automatically learns the attention between the same type of data elements in two subgraphs, enabling it to better identify the correspondence between the two subgraphs.

[0055] exist Figure 1 In the specific implementation shown, the multi-packet efficient matching and fusion method based on connected graphs further includes:

[0056] Step S105: Based on the most certain match, obtain the objects corresponding to other most certain matches that are geometrically consistent and adjacent to the most certain match.

[0057] In this implementation, if a pair of matches represents the same object, it meets the geometric consistency requirement. Near the most certain match, the matching results of uncertain objects are checked to ensure that no object is missed in the low-precision single-package map, guaranteeing the completeness of the matching.

[0058] In one specific embodiment of this application, obtaining objects corresponding to other most confident matches that conform to geometric consistency and are adjacent to the most confident match based on the most confident match includes: selecting objects corresponding to other most confident matches that conform to geometric consistency and are adjacent to the most confident match multiple times based on the most confident match, using a first correlation threshold in the first level and a second correlation threshold in the second level.

[0059] In this embodiment, after each batch of most certain matches is found, the most certain matches among the uncertain objects in the vicinity of these most certain matches are searched sequentially. Two levels of filtering and matching of uncertain objects ensure the accuracy of the matching.

[0060] exist Figure 1 In the specific implementation shown, the multi-packet efficient matching and fusion method based on connected graphs further includes:

[0061] Step S106: Align the objects that are most confidently matched in every two low-precision single-packet maps one by one to obtain a high-precision single-packet map.

[0062] In this embodiment, through multiple iterations of pose graph optimization, matching objects from every two low-precision single-packet maps are combined to obtain high-precision single-packet maps. This improves the accuracy of the maps.

[0063] exist Figure 1 In the specific implementation shown, the multi-packet efficient matching and fusion method based on connected graphs further includes:

[0064] Step S107: Merge and optimize the high-precision single-package maps to obtain a high-precision map.

[0065] In this embodiment, the map accuracy is improved through fusion optimization.

[0066] This application uses a K+1 matching method, which saves computational complexity while ensuring matching stability. First, the same center point in every pair of K adjacent low-precision single-packet maps is connected to ensure a complete connected graph. To prevent further deviations in subsequent matching if two adjacent low-precision single-packet maps do not match well, each low-precision single-packet map is matched with the other low-precision single-packet map with the largest overlapping area to ensure matching stability.

[0067] Figure 2 This paper illustrates a specific implementation of a multi-packet efficient matching and fusion device based on a connected graph, as described in this application. Figure 2 In the specific implementation shown, the multi-packet efficient matching and fusion device based on connected graphs mainly includes:

[0068] Module 201 is used to obtain a low-precision single-packet map of N non-overlapping data elements in the same task, where N is a natural number greater than 1.

[0069] Module 202 is used to generate a sub-map with a fixed range in each low-precision single-package map, with randomly selected data elements as the center points.

[0070] Module 203 is used to automatically match the two subgraphs corresponding to the same center point in each of the K low-precision single-packet maps that pass through the same center point, and sequentially connect the same center point in the K low-precision single-packet maps to obtain the matching score corresponding to different data elements between each two subgraphs in the complete connected graph. The module that determines the object corresponding to the most certain matching that meets geometric consistency is determined by the matching score, where K is a natural number less than N and greater than 0.

[0071] Module 204 is used to automatically match the two sub-maps that each low-precision single-packet map passes through with the same center point in the other low-precision single-packet map with the largest overlapping area, obtain the matching score corresponding to different data elements between the two sub-maps, and determine the object corresponding to the most certain match that meets geometric consistency through the matching score.

[0072] Module 205 is used to obtain the objects corresponding to other geometrically consistent most certain matches that are adjacent to the most certain match, based on the most certain match.

[0073] Module 206 is used to align the objects that are most confidently matched between every two low-precision single-packet maps one by one to obtain a high-precision single-packet map.

[0074] Module 207 is used to merge and optimize high-precision single-package maps to obtain high-precision maps.

[0075] In this embodiment, by performing loop-based segmentation on the low-precision single-packet map containing backtracking paths, each object has only one corresponding data element in a single low-precision single-packet map, facilitating subsequent matching. Dividing the low-precision single-packet map into multiple sub-maps for matching improves efficiency and accuracy, and solves the problem of high costs associated with high-precision equipment. The most certain matching method ensures that this solution does not rely on the uniqueness of map elements. Connecting the same center point in every K adjacent low-precision single-packet maps to form a connected graph saves computational power. Automatic matching of the two sub-maps through which each low-precision single-packet map passes with the same center point from the other low-precision single-packet map with the largest overlap area ensures the stability of the matching.

[0076] The multi-packet efficient matching and fusion apparatus based on connected graphs provided in this application can be used to execute the multi-packet efficient matching and fusion method based on connected graphs described in any of the above embodiments. Its implementation principle and technical effect are similar, and will not be repeated here.

[0077] In one specific embodiment of this application, the functional modules of the multi-packet high-efficiency matching and fusion device based on connected graphs can be directly in hardware, in software modules executed by a processor, or in a combination of both.

[0078] Software modules may reside in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, registers, hard disks, removable disks, CD-ROMs, or any other form of storage medium known in this art. An exemplary storage medium is coupled to the processor, enabling the processor to read information from and write information to the storage medium.

[0079] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic, discrete hardware components, or any combination thereof. A general-purpose processor can be a microprocessor, but alternatively, it can be any conventional processor, controller, microcontroller, or state machine. The processor can also be implemented as a combination of computing devices, such as a combination of a DSP and a microprocessor, multiple microprocessors, one or more microprocessors incorporating a DSP core, or any other such configuration. Alternatively, the storage medium can be integrated with the processor. The processor and storage medium can reside in an ASIC. The ASIC can reside in the user terminal. Alternatively, the processor and storage medium can reside as discrete components in the user terminal.

[0080] In another specific embodiment of this application, a computer-readable storage medium stores computer instructions that are operated to perform the multi-packet efficient matching and fusion method based on a connected graph in any embodiment.

[0081] In another specific embodiment of this application, a computer device includes a processor and a memory, the memory storing computer instructions that are operated to perform the multi-packet efficient matching and fusion method based on a connected graph in any embodiment.

[0082] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0083] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0084] The above description is merely an embodiment of this application and does not limit the patent scope of this application. Any equivalent structural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.

Claims

1. A multi-packet efficient matching and fusion method based on connected graphs, characterized in that, include: Get N low-precision single-packet maps of the same task that do not have overlapping data elements, where N is a natural number greater than 1; In each of the low-precision single-package maps, a sub-map with a fixed range is generated with a randomly selected data element as the center point. Each of the low-precision single-package maps includes multiple sub-maps, and there are overlapping areas between the multiple sub-maps. In the K low-precision single-packet maps that pass through the same center point, the two subgraphs corresponding to the same center point in each of the two adjacent low-precision single-packet maps are automatically matched, and the same center point in the K low-precision single-packet maps is connected in sequence to obtain the matching score corresponding to different data elements between each two subgraphs in the complete connected graph. The object corresponding to the most certain match that meets the geometric consistency is determined by the matching score, where K is a natural number less than N and greater than 0. Each low-precision single-packet map and the other low-precision single-packet map with the largest overlapping area are re-matched through two sub-maps to obtain matching scores for different data elements between the two sub-maps. The objects corresponding to the most certain matching that meet geometric consistency are determined by the matching scores. Based on the most certain match, obtain the objects corresponding to other most certain matches that are adjacent to the most certain match and conform to the geometric consistency. Align the objects corresponding to the most certain match in every two low-precision single-packet maps of the N low-precision single-packet maps that do not contain overlapping data elements one by one to obtain a high-precision single-packet map. The high-precision single-package maps are merged and optimized to obtain a high-precision map.

2. The efficient multi-packet matching and fusion method based on connected graphs as described in claim 1, characterized in that, The step of re-matching each of the low-precision single-packet maps with the two sub-maps through which the same center point passes includes: Based on the number of sub-maps traversed by the same center point in the two low-precision single-packet maps, select the two sub-maps with the largest overlapping area in the two low-precision single-packet maps. Simultaneously extract features from different data elements between the two subgraphs with the largest overlapping regions to obtain feature information corresponding to the two subgraphs; The feature information corresponding to the two subgraphs is input into the attention map neural network for automatic matching, and the matching score corresponding to different data elements between the two subgraphs is obtained.

3. The efficient multi-packet matching and fusion method based on connected graphs as described in claim 2, characterized in that, The step of simultaneously extracting features from different data elements between the two sub-graphs with the largest overlapping regions to obtain their respective feature information includes: Based on the differences in the surrounding elements and their own attributes of different data elements, feature extraction is performed on different data elements in the same subgraph. When a data element belongs to the first category, feature extraction is performed on its surrounding elements and its own attributes to obtain corresponding semantic feature information and perceptual feature information. The semantic feature information and the perceptual feature information are superimposed to obtain the overall feature information corresponding to the data element. When another data element belongs to the second category, its own attributes are extracted to obtain attribute feature information, and then feature information corresponding to different types of data elements in the same subgraph is obtained.

4. The efficient multi-packet matching and fusion method based on connected graphs as described in claim 2, characterized in that, The step of inputting the feature information corresponding to the two subgraphs into an attention map neural network for automatic matching to obtain matching scores for different data elements between the two subgraphs includes: The attention graph neural network automatically learns attention for the feature information of data elements corresponding to two subgraphs, and simultaneously learns attention for the feature information of different data elements in the same subgraph, to obtain matching scores for different data elements between the two subgraphs.

5. The efficient multi-packet matching and fusion method based on connected graphs as described in claim 1, characterized in that, The process of obtaining N low-precision single-packet maps of non-overlapping data elements from the same task includes: If the low-precision single-package map contains backtracking paths, it is split at the backtracking paths to obtain two low-precision single-package maps that do not contain overlapping data elements.

6. The efficient multi-packet matching and fusion method based on connected graphs as described in claim 1, characterized in that, The step of obtaining other objects corresponding to the geometrically consistent most certain matches that are adjacent to the most certain match includes: Based on the most certain match, using the first correlation threshold in the first level and the second correlation threshold in the second level, the objects corresponding to other most certain matches that conform to the geometric consistency and are adjacent to the most certain match are selected multiple times.

7. The efficient multi-packet matching and fusion method based on connected graphs as described in claim 1, characterized in that, The step of determining the object corresponding to the most certain match that meets geometric consistency through the matching score includes: When the matching score is consistent with the benchmark threshold, the data element corresponding to the matching score meets the geometric consistency. Furthermore, when the proportion of the number of matching scores that are consistent with the benchmark threshold is greater than the preset threshold, the data element corresponding to the matching score is the object that is most confidently matched.

8. A multi-packet efficient matching and fusion device based on connected graphs, characterized in that, include: A module for obtaining a low-precision single-packet map of N non-overlapping data elements in the same task, where N is a natural number greater than 1; A module for generating a corresponding fixed-range sub-map in each of the low-precision single-package maps, with randomly selected data elements as the center point, wherein each of the low-precision single-package maps includes multiple sub-maps, and there are overlapping areas between the multiple sub-maps; In the K low-precision single-packet maps that pass through the same center point, the two subgraphs corresponding to the same center point in each of the two adjacent low-precision single-packet maps are automatically matched, and the same center point in the K low-precision single-packet maps is connected in sequence to obtain the matching score corresponding to different data elements between each two subgraphs in the complete connected graph. The module corresponding to the object with the most certain matching that meets geometric consistency is determined by the matching score, where K is a natural number less than N and greater than 0. A module for rematching the two sub-maps through which each of the low-precision single-package maps passes and the other low-precision single-package map with the largest overlapping area passes, to obtain matching scores for different data elements between the two sub-maps, and to determine the object corresponding to the most certain match that meets geometric consistency through the matching scores; A module for obtaining, based on the most certain match, other objects that conform to the geometric consistency and are adjacent to the most certain match; A module for aligning the objects corresponding to the most certain match in every two low-precision single-packet maps of the N non-overlapping data elements one by one to obtain a high-precision single-packet map. This module is used to fuse and optimize the high-precision single-package maps to obtain a high-precision map.

9. A computer-readable storage medium storing computer instructions, characterized in that, The computer instructions are operated to perform the efficient multi-packet matching and fusion method based on connected graphs as described in any one of claims 1-7.

10. A computer device comprising a processor and a memory storing computer instructions, wherein the processor operates the computer instructions to perform the multi-packet efficient matching and fusion method based on a connected graph according to any one of claims 1-7.

Images