A Geographic Scene Instantiation Modeling Method

Abstract

The present application is applicable to the field of geographic information processing technology, and provides a geographic scene instantiation modeling method, including: dividing a research area into multiple grid units, and taking each grid unit as a geographic scene; constructing a geographic scene ontology model; the geographic scene ontology model includes multiple geographic scene nodes and multiple geographic element nodes, the multiple geographic scene nodes correspond one-to-one to the multiple grid units, each geographic scene node represents a corresponding geographic scene, the multiple geographic element nodes correspond one-to-one to the multiple entities in the research area, and each geographic scene node is provided with a relationship edge with the geographic element node corresponding to the entity located in the corresponding grid unit; based on a dynamically adjusted similarity threshold, the geographic scene nodes in the geographic scene ontology model are merged multiple times, and the geographic scene ontology model after multiple merging processes is used as the geographic scene instantiation model of the research area. The present application can realize geographic scene instantiation modeling.

Description

Technical Field

[0001] The present application belongs to the field of geographic information processing technology, and in particular, relates to a geographic scene instantiation modeling method. Background Art

[0002] Maps are an important tool for humans to understand, describe and express the real world. Geographical elements are map elements with geometric shapes that can represent real entities, such as settlements, transportation networks, rivers and lakes. Current maps mainly express entities based on layer overlays, artificially severing the inherent relationships between entities, violating human cognitive habits of the real geographical world, and facing many challenges in serving diversified, personalized and precise human-oriented application needs. Geographic Scenario is a regional complex with specific structure, function and semantics. It is the basic unit for humans to understand the real geographical world (such as " University"," Mountain scenic area", etc.), which reflects the systematic and structural cognition of human beings on the relationship between entities, and conforms to human beings' cognitive habits of the geographical world. The production of new knowledge service maps (referred to as knowledge maps) based on geographical scenarios is an urgent need for new application scenarios such as digital twins and autonomous driving in the era of big data, and is an effective way to realize the transformation of the map service paradigm from digital information services to intelligent knowledge services.

[0003] At present, a lot of related work has been carried out around geographic scenes. Some scholars recognize and analyze the structure of geographic scenes from the perspective of holism and reductionism, and believe that geographic scenes contain six types of elements: people, events, things, phenomena, time, and place. They also characterize the entities in geographic scenes from six aspects: geometric form, spatial position, semantic description, attribute characteristics, evolution process, and element relationship, forming a systematic description of the concept of geographic scenes and constructing geographic scene data models and ontology expression frameworks. Other scholars advocate modeling specific scenes in the real world by connecting entities and establishing entity-space-time association networks, such as farmland scenes composed of fields, wells, and pipelines. On this basis, some scholars have introduced the idea of ​​multi-granularity spatio-temporal objects, using spatio-temporal objects of different granularities to represent things in the geographical world that can be distinguished from each other. These spatio-temporal objects can be split or merged into spatio-temporal objects of other granularities. The smallest granularity spatio-temporal object is the entity. The smallest granularity spatio-temporal object is merged upward to form a coarser-grained spatio-temporal object (i.e., geographical scene). The coarse-grained spatio-temporal objects continue to merge upward to form a coarser-grained spatio-temporal object (i.e., a more macroscopic geographical scene). Through layer-by-layer modeling, a multi-scale description of the real-world geographical scene is formed to realize the practical application of the geographical scene.

[0004] However, current research on geographic scenarios still mainly revolves around conceptual modeling, focusing on the composition structure and data model of geographic scenarios. A relatively complete technical method has not yet been formed for the instantiation modeling of geographic scenarios. Summary of the invention

[0005] The embodiment of the present application provides a geographic scene instantiation modeling method, which can solve the problem that geographic scene instantiation modeling cannot be implemented.

[0006] The present application provides a geographic scene instantiation modeling method, including:

[0007] Step 1: divide the study area into multiple grid units, and use each grid unit as a geographical scene;

[0008] Step 2, constructing a geographic scene ontology model; the geographic scene ontology model includes multiple geographic scene nodes and multiple geographic element nodes, the multiple geographic scene nodes correspond one-to-one to multiple grid cells, each geographic scene node represents a corresponding geographic scene, the multiple geographic element nodes correspond one-to-one to multiple entities in the study area, and a relationship edge is set between each geographic scene node and the geographic element node corresponding to the entity located in the corresponding grid cell;

[0009] Step 3: Based on the dynamically adjusted similarity threshold, the geographic scene nodes in the geographic scene ontology model are merged multiple times, and the geographic scene ontology model after multiple merging processes is used as the geographic scene instantiation model of the study area.

[0010] Optionally, all geographic scene nodes in the constructed geographic scene ontology model are level 0 geographic scene nodes. After the geographic scene nodes in the geographic scene ontology model are merged for the nth time based on the dynamically adjusted similarity threshold, the highest level geographic scene node in the geographic scene ontology model is the level n geographic scene node. , Indicates the total number of levels of the preset geographic scene nodes.

[0011] Optionally, performing an n-th merging process on the geographic scene nodes in the geographic scene ontology model based on a dynamically adjusted similarity threshold includes:

[0012] Step 3.1, taking any n-1 level geographic scene node in the geographic scene ontology model as the current traversal node;

[0013] Step 3.2: Based on the set similarity threshold, the currently traversed node is used as the starting point to search for unvisited nodes. n -1 level geographic scene nodes are merged to obtain the corresponding nodes of the current traversal nThe geographical scene node of the level, and determine the corresponding geographical scene node in the geographical scene ontology model that is subordinate to the current traversal node n Level geographic scene node n -Level 1 geographic scene node; unvisited n -Level 1 geographic scene nodes are not subordinate to any n Level geographic scene node n -Level 1 geographic scene node;

[0014] Step 3.3, select any n-1 level geographic scene node from the n-1 level geographic scene nodes that have not been visited as a new current traversal node, and use the new current traversal node as the current traversal node in step 3.2 to return to execute step 3.2 until all n-1 level geographic scene nodes in the geographic scene ontology model have subordinate n level geographic scene nodes;

[0015] Step 3.4, determine whether the ratio of the number of level n geographic scene nodes to the number of level n-1 geographic scene nodes in the geographic scene ontology model is greater than the merge ratio threshold. If so, lower the similarity threshold, delete all level n geographic scene nodes, and return to execute step 3.1 until the ratio of the number of level n geographic scene nodes to the number of level n-1 geographic scene nodes in the geographic scene ontology model is less than or equal to the merge ratio threshold.

[0016] Optionally, the calculation formula for the merge ratio threshold is:

[0017]

[0018] in, Indicates the merge ratio threshold, represents the proportionality coefficient, represents the magnification factor, .

[0019] Optionally, step 3.2 includes:

[0020] Based on the geographical scene corresponding to the current traversal node, determine the adjacent n-1-level geographical scene nodes of the current traversal node from the n-1-level geographical scene nodes that have not been visited; the geographical scene corresponding to the adjacent n-1-level geographical scene nodes of the current traversal node is adjacent to the geographical scene corresponding to the current traversal node;

[0021] Calculate the current traversal node and each adjacent n -1 similarity of geographic scene nodes, and the adjacent nodes whose similarity to the currently traversed node is greater than the similarity threshold n -1-level geographic scene node and current traversal node are merged to generate the corresponding n Level geographic scene node;

[0022] The current traversal node and its adjacent nodes whose similarity to the current traversal node is greater than the similarity threshold are n -Level 1 geographic scene node as a subordinate to the current traversal node n Level geographic scene node n -Level 1 geographic scene node;

[0023] The geographical scenes corresponding to all n-1-level geographical scene nodes belonging to the n-level geographical scene node are used as the geographical scenes corresponding to the n-level geographical scene node, and based on the geographical scenes corresponding to the n-level geographical scene node, the adjacent n-1-level geographical scene nodes of the n-level geographical scene node are determined from the n-1-level geographical scene nodes that have not been visited; the geographical scenes corresponding to the determined adjacent n-1-level geographical scene nodes are adjacent to the geographical scenes corresponding to the n-level geographical scene node;

[0024] Determine whether there is an adjacent n-1 level geographic scene node whose similarity with the n-level geographic scene node is greater than a similarity threshold among the adjacent n-1 level geographic scene nodes of the n-level geographic scene node; if so, take the n-1 level geographic scene node whose similarity with the n-level geographic scene node is greater than the similarity threshold as the n-1 level geographic scene node subordinate to the n-level geographic scene node;

[0025] Return the steps of taking the geographic scenes corresponding to all n-1 level geographic scene nodes belonging to the n-level geographic scene node as the geographic scene corresponding to the n-level geographic scene node, until there are no adjacent n-1 level geographic scene nodes of the n-level geographic scene node among the unvisited n-1 level geographic scene nodes, or, there are no adjacent n-1 level geographic scene nodes of the n-level geographic scene node whose similarity with the n-level geographic scene node is greater than a similarity threshold.

[0026] Optionally, the similarity between the current traversal node and each adjacent n-1 level geographic scene node is calculated, including:

[0027] The similarity between the current traversal node and each adjacent n-1 level geographic scene node is calculated by the similarity formula; the similarity formula is:

[0028] ;

[0029] in, Indicates the current traversal node The similarity with the jth adjacent n-1 level geographic scene node, represents adjustable parameters, represents the discount factor, Indicates the execution round of the merge operation. represents the area of ​​the geographic scene corresponding to the jth adjacent n-1 level geographic scene node, represents the sum of the areas of all geographic scenes, Indicates the number of entities contained in the geographic scene corresponding to the jth adjacent n-1 level geographic scene node, Represents the sum of the number of entities contained in all geographic scenes, Indicates the current traversal node The feature vector of the corresponding geographic scene, Represents the feature vector of the geographic scene corresponding to the jth adjacent n-1 level geographic scene node.

[0030] Optionally, the geographic scenario instantiation modeling method further includes:

[0031] A relationship edge is set between the n-level geographic scene node corresponding to the current traversal node and the n-1-level geographic scene node subordinate to the n-level geographic scene node.

[0032] Optionally, the multiple entities include buildings, roads, and POI points.

[0033] The above solution of the present application has the following beneficial effects:

[0034] In an embodiment of the present application, the study area is divided into multiple geographic scenes, and the correspondence between geographic scene nodes and geographic scenes, and the correspondence between geographic element nodes and entities are set based on the divided geographic scenes and entities in the study area. Then, the corresponding relationship edges are set based on the association between the geographic scene nodes and the geographic element nodes to complete the construction of the geographic scene ontology model. Finally, based on a dynamically adjustable similarity threshold, the geographic scene nodes in the geographic scene ontology model are merged multiple times to realize geographic scene instantiation modeling to meet the actual application needs related to geographic scenes.

[0035] Other beneficial effects of the present application will be described in detail in the subsequent specific implementation section. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0037] Figure 1 A flowchart of a geographic scene instantiation modeling method provided in an embodiment of the present application;

[0038] Figure 2 A schematic diagram of merging level 0 geographic scene nodes provided in an embodiment of the present application;

[0039] Figure 3 This is a schematic diagram of a geographic scene instantiation model in an example of the present application. DETAILED DESCRIPTION

[0040] In the following description, specific details such as specific system structures, technologies, etc. are provided for the purpose of illustration rather than limitation, so as to provide a thorough understanding of the embodiments of the present application. However, it should be clear to those skilled in the art that the present application may also be implemented in other embodiments without these specific details. In other cases, detailed descriptions of well-known systems, devices, circuits, and methods are omitted to prevent unnecessary details from obstructing the description of the present application.

[0041] It should be understood that when used in the present specification and the appended claims, the term "comprising" indicates the presence of described features, wholes, steps, operations, elements and / or components, but does not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components and / or combinations thereof.

[0042] It should also be understood that the term “and / or” used in the specification and appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0043] As used in the specification and appended claims of this application, the term "if" can be interpreted as "when" or "uponce" or "in response to determining" or "in response to detecting", depending on the context. Similarly, the phrase "if it is determined" or "if [described condition or event] is detected" can be interpreted as meaning "uponce it is determined" or "in response to determining" or "uponce [described condition or event] is detected" or "in response to detecting [described condition or event]", depending on the context.

[0044] In addition, in the description of the present application specification and the appended claims, the terms "first", "second", "third", etc. are only used to distinguish the descriptions and cannot be understood as indicating or implying relative importance.

[0045] References to "one embodiment" or "some embodiments" etc. described in the specification of this application mean that one or more embodiments of the present application include specific features, structures or characteristics described in conjunction with the embodiment. Therefore, the statements "in one embodiment", "in some embodiments", "in some other embodiments", "in some other embodiments", etc. that appear in different places in this specification do not necessarily refer to the same embodiment, but mean "one or more but not all embodiments", unless otherwise specifically emphasized in other ways. The terms "including", "comprising", "having" and their variations all mean "including but not limited to", unless otherwise specifically emphasized in other ways.

[0046] In view of the problem that it is currently impossible to implement geographic scene instantiation modeling, the embodiment of the present application provides a geographic scene instantiation modeling method, which divides the study area into multiple geographic scenes, and sets the correspondence between geographic scene nodes and geographic scenes, and the correspondence between geographic element nodes and entities based on the divided geographic scenes and entities in the study area, and then sets the corresponding relationship edges based on the association between geographic scene nodes and geographic element nodes to complete the construction of the geographic scene ontology model, and finally merges the geographic scene nodes in the geographic scene ontology model multiple times based on a dynamically adjustable similarity threshold, so as to realize geographic scene instantiation modeling to meet the actual application needs related to geographic scenes. The modeling results can provide underlying model support for the new generation of knowledge service maps based on geographic scenes, and provide intelligent map knowledge services for new application scenarios such as digital twins and autonomous driving.

[0047] The geographic scene instantiation modeling method provided in this application is exemplarily described below in conjunction with specific embodiments.

[0048] like Figure 1 As shown, the geographic scene instantiation modeling method provided in the embodiment of the present application includes the following steps:

[0049] Step 1: Divide the study area into multiple grid units, and use each grid unit as a geographical scene.

[0050] The above-mentioned study area is an area where geographic scene instantiation modeling is required. In some embodiments of the present application, the entire study area can be divided into regular grids. The grid unit size can be 200m×200m. A grid unit is a geographic scene. The geographic scene is a regional complex that contains a large number of entities (corresponding to the geographic elements in the map). The geometric shape of the grid unit is the geometric shape of the geographic scene, and all entities located inside the grid unit belong to the geographic scene represented by the grid unit. Count the number of different types of entities contained in the geographic scene and represent them with a vector (i.e., a feature vector). In the subsequent calculation process, this feature vector can be used to describe the geographic scene.

[0051] It should be noted that, in order to facilitate subsequent merging operations, adjacent grid cells are set to partially overlap, so as to facilitate rapid query of adjacent geographical scenes, and the existence of partial overlap of adjacent geographical scenes is also consistent with the situation in the real world.

[0052] Step 2: Build a geographic scene ontology model.

[0053] In some embodiments of the present application, a geographic scene ontology model may be constructed based on a graph structure. Specifically, a geographic scene node may be used to represent a geographic scene, a geographic element node may be used to represent an entity within a research area, and a relationship edge may be used to connect the geographic scene node with the geographic element node within the scene range (i.e., within the grid unit corresponding to the geographic scene node) to form a description of the geographic scene.

[0054] Based on the above construction principles, the constructed geographic scene ontology model includes multiple geographic scene nodes and multiple geographic element nodes. Multiple geographic scene nodes correspond to multiple grid cells one by one. Each geographic scene node represents the corresponding geographic scene. Multiple geographic element nodes correspond to multiple entities in the study area one by one. There is a relationship edge between each geographic scene node and the geographic element node corresponding to the entity in its corresponding grid cell. The above multiple entities include buildings, roads, and points of interest (POI) (such as xx hotel, xx gas station, xx supermarket, xx food, etc.).

[0055] Among them, the data description of the geographic scene node is shown in Table 1, which describes the unique identification code, scene name, level label, time label, geometric shape and other information of the geographic scene, as well as the information of the entities it contains, represented by the type and quantity of the entity; the data description of the geographic element node is shown in Table 2, which describes the unique identification code, name, type, time label, geometric shape and other information of the entity (corresponding to the geographic element on the map).

[0056] Table 1 Data description of geographic scene nodes

[0057]

[0058] Table 2 Data description of geographic element nodes

[0059]

[0060] In Table 1 and Table 2 above, int represents integer type, string represents text type, list[string] represents text list type, list[int] represents integer list type, geometry represents spatial data type, EPSG:4326 represents geographic coordinate system, POLYGON represents polygon feature, and POINT represents point feature.

[0061] Step 3: Based on the dynamically adjusted similarity threshold, the geographic scene nodes in the geographic scene ontology model are merged multiple times, and the geographic scene ontology model after multiple merging processes is used as the geographic scene instantiation model of the study area.

[0062] It should be noted that all geographic scene nodes in the constructed geographic scene ontology model are level 0 geographic scene nodes; multiple adjacent geographic scene nodes can be merged with each other to generate a new macro-geographic scene node containing multiple geographic scene nodes. After the geographic scene nodes in the geographic scene ontology model are merged for the nth time based on the dynamically adjusted similarity threshold, the highest-level geographic scene node in the geographic scene ontology model is the level n geographic scene node. , Indicates the total number of levels of the preset geographic scene nodes. For example, Can be set to 20.

[0063] That is, in the geographic scene ontology model, the hierarchy of geographic scene nodes is calculated from level 0. The level 0 geographic scene node represents the smallest unit geographic scene, and its subordinate nodes are geographic element nodes. The subordinate nodes of the level 1 geographic scene node are level 0 geographic scene nodes, which contain all the geographic element nodes contained in all its subordinate level 0 scene nodes. And so on, through the layer-by-layer merging of geographic scene nodes, higher-level geographic scene nodes are formed until a geographic scene node with the highest level containing all data information is generated.

[0064] Among them, when the geographic scene nodes in the geographic scene ontology model are merged for the nth time based on the dynamically adjusted similarity threshold, the n-1-level geographic scene nodes that are not subordinate to any n-level geographic scene nodes are marked as unvisited n-1-level geographic scene nodes. Of course, as long as the unvisited n-1-level geographic scene nodes are determined to be subordinate to any n-level geographic scene nodes, the n-1-level geographic scene nodes will be marked as visited n-1-level geographic scene nodes. That is, the unvisited n-1-level geographic scene nodes are n-1-level geographic scene nodes that are not subordinate to any n-level geographic scene nodes. Exemplarily, when the first merging process is performed, the 0-level geographic scene nodes that are not subordinate to any 1-level geographic scene nodes are marked as unvisited 0-level geographic scene nodes.

[0065] In some embodiments of the present application, a specific implementation method of performing the n-th merging process on the geographic scene nodes in the geographic scene ontology model based on the dynamically adjusted similarity threshold includes the following steps 3.1 to 3.4:

[0066] Step 3.1, taking any n-1 level geographic scene node in the geographic scene ontology model as the current traversal node.

[0067] Step 3.2: Based on the set similarity threshold, the currently traversed node is used as the starting point to search for unvisited nodes. n -1 level geographic scene nodes are merged to obtain the corresponding nodes of the current traversal nThe geographical scene node of the level, and determine the corresponding geographical scene node in the geographical scene ontology model that is subordinate to the current traversal node n Level geographic scene node n -Level 1 geographic scene node.

[0068] In some embodiments of the present application, the unvisited nodes may be sorted based on the similarity of the adjacent geographic scene nodes. n -1 level geographic scene nodes are merged (specifically, adjacent geographic scene nodes whose similarity is greater than the similarity threshold can be merged), and the corresponding nodes in the geographic scene ontology model that are subordinate to the current traversal node are determined. n Level geographic scene node n -Level 1 geographic scene node.

[0069] Step 3.3, select any n-1 level geographic scene node from the n-1 level geographic scene nodes that have not been visited as the new current traversal node, and return the new current traversal node as the current traversal node in step 3.2 to execute step 3.2 until all n-1 level geographic scene nodes in the geographic scene ontology model have subordinate n level geographic scene nodes.

[0070] Step 3.4, determine whether the ratio of the number of level n geographic scene nodes to the number of level n-1 geographic scene nodes in the geographic scene ontology model is greater than the merge ratio threshold. If so, lower the similarity threshold, delete all level n geographic scene nodes, and return to execute step 3.1 until the ratio of the number of level n geographic scene nodes to the number of level n-1 geographic scene nodes in the geographic scene ontology model is less than or equal to the merge ratio threshold.

[0071] It should be noted that since the similarity between adjacent geographic scene nodes decreases as the level increases, the higher the level of geographic scene nodes, the more difficult it is to merge each other, which has a negative impact on the smooth switching between geographic scenes of different scales and leads to scale redundancy. Therefore, in order to ensure the smoothness and continuity of scale transformation between geographic scenes of different levels (different scales) and effectively avoid hierarchical redundancy (avoid the situation where the structures of geographic scenes at different levels are very similar), this application dynamically adjusts the similarity threshold by controlling the merging ratio of geographic scenes, thereby achieving a smooth transition of scale transformation between geographic scenes of different levels and generating a multi-scale geographic scene instantiation model. Among them, the calculation formula for the above-mentioned merging ratio threshold is:

[0072]

[0073] in, Indicates the merge ratio threshold. When the ratio of the number of level n geographic scene nodes to the number of level n-1 geographic scene nodes is greater than , then appropriately lower the similarity threshold, delete all the determined n-level geographic scene nodes (it can be understood that after deleting the n-level geographic scene node, the subordinate relationship between the n-level geographic scene node and its subordinate n-1-level geographic scene nodes is also deleted), and re-merge the geographic scenes until the ratio of the number of higher-level geographic scenes generated by the re-merger (i.e., the number of n-level geographic scene nodes) to the number of current-level geographic scenes (i.e., the number of n-1-level geographic scene nodes) is less than, represents the proportionality coefficient, represents the magnification factor, , and Can be set according to actual situation.

[0074] So, based on By controlling the merging ratio of geographic scenes and dynamically adjusting the similarity threshold, the similarity threshold is continuously reduced as the scene level increases, and the difficulty of merging adjacent scenes is gradually reduced as the scene level increases. This can effectively ensure the smoothness and continuity of the scale transformation of geographic scenes, and ensure the effectiveness of automatically merged geographic scenes of different scales.

[0075] The specific implementation method of the above step 3.2 is exemplified below.

[0076] Specifically, in step 3.2 above, based on the set similarity threshold, the currently traversed node is used as the starting point to search for unvisited nodes. n -1 level geographic scene nodes are merged to obtain the corresponding nodes of the current traversal n The geographical scene node of the level, and determine the corresponding geographical scene node in the geographical scene ontology model that is subordinate to the current traversal node n Level geographic scene node n -The specific implementation of the level 1 geographic scene node includes the following steps:

[0077] Step 3.21, based on the geographical scene corresponding to the current traversal node, determine the adjacent n-1-level geographical scene nodes of the current traversal node from the n-1-level geographical scene nodes that have not been visited. The geographical scene corresponding to the adjacent n-1-level geographical scene nodes of the current traversal node is adjacent to the geographical scene corresponding to the current traversal node. This adjacency can be the eight neighborhoods when the grid is divided, or it can be understood as the overlap of grid units. That is, for a certain geographical scene, the geographical scene corresponding to its eight neighborhoods can be regarded as its adjacent geographical scene, or the geographical scene with overlapping areas can be regarded as its adjacent geographical scene.

[0078] Step 3.22, calculate the current traversal node and each adjacent n -1 similarity of geographic scene nodes, and the adjacent nodes whose similarity to the currently traversed node is greater than the similarity thresholdn -1-level geographic scene node and current traversal node are merged to generate the corresponding n Level geographic scene node.

[0079] Step 3.23, the current traversal node and the adjacent nodes whose similarity with the current traversal node is greater than the similarity threshold n -Level 1 geographic scene node as a subordinate to the current traversal node n Level geographic scene node n -Level 1 geographic scene node.

[0080] Step 3.24, the geographical scenes corresponding to all n-1 level geographical scene nodes belonging to the n level geographical scene node are used as the geographical scenes corresponding to the n level geographical scene node.

[0081] In some embodiments of the present application, the similarity between the current traversal node and each adjacent n-1 level geographic scene node can be calculated by a similarity formula. The similarity formula is:

[0082] ;

[0083] in, Indicates the current traversal node The similarity with the jth adjacent n-1 level geographic scene node, represents adjustable parameters, represents the discount factor, The value range of is [0, 1], Used to limit the merge expansion speed of geographical scenarios, Indicates the execution round of the merge operation (i.e. n), represents the area of ​​the geographic scene corresponding to the jth adjacent n-1 level geographic scene node, represents the sum of the areas of all geographic scenes, Indicates the number of entities contained in the geographic scene corresponding to the jth adjacent n-1 level geographic scene node, Represents the sum of the number of entities contained in all geographic scenes, Indicates the current traversal node The feature vector of the corresponding geographic scene, The feature vector representing the geographic scene corresponding to the j-th adjacent n-1-level geographic scene node. The feature vector is used to represent the number of different types of entities contained in the corresponding geographic scene. The number of different types of entities is represented by a vector.

[0084] It is worth mentioning that the introduction of the area of ​​the geographical scene and the number of entities it contains makes it easier to merge scenes with smaller areas and more difficult to merge scenes with more entities, thereby ensuring that geographical scenes with more significant features are retained and highlighted during the merging process. In this way, based on the similarity formula, while maintaining a relative balance in scale between geographical scenes, it is possible to automatically merge adjacent geographical scenes at the same level and generate geographical scenes at a higher level, effectively avoiding the situation where the newly generated geographical scenes are too large or too small, and retaining and highlighting geographical scenes with more significant features.

[0085] For example, Figure 2 As shown, for the geographic scene node named level 0 scene 0 (representing Figure 2 The 0th level scene 0 (i.e., the geographic scene) and the geographic scene node named 0th level scene 1 (representing Figure 2 0 level 1 scene) to generate a geographic scene node named level 1 scene 0 (indicating Figure 2 Medium Level 1 Scene 0). Among them, Figure 2 The geographic scene node named level 0 scene 0 and the entity located in its corresponding geographic scene ( Figure 2 There are relationship edges between the geographical element nodes corresponding to xx supermarket, xx gas station, etc. Figure 2 The geographic scene node named level 0 scene 1 and the entity located in its corresponding geographic scene ( Figure 2 There are relationship edges between the geographical element nodes corresponding to xx hotel, xx food, xx supermarket, etc. Figure 2 The three nine-square grids in the upper middle are schematic diagrams of gridded grid units. The arrow pointing to xx supermarket is used to indicate that xx supermarket is in the grid unit of the 3rd row and 1st column. The arrow pointing to level 0, No. 0 scene is used to indicate that the grid unit corresponding to level 0, No. 0 scene is the grid unit of the 3rd row and 1st column. The arrow pointing to level 1, No. 0 scene is used to indicate that the grid unit corresponding to level 1, No. 0 scene is the grid unit composed of the grid unit of the 3rd row and 1st column and the grid unit of the 3rd row and 2nd column.

[0086] Step 3.25, based on the geographic scene corresponding to the n-level geographic scene node (the n-level geographic scene node corresponding to the current traversal node), determine the adjacent n-1-level geographic scene nodes of the n-level geographic scene node from the n-1-level geographic scene nodes that have not been visited.

[0087] Among them, the geographical scene corresponding to the determined adjacent n-1-level geographical scene node is adjacent to the geographical scene corresponding to the n-level geographical scene node. The adjacent here is the same as the adjacent in step 3.21, which can be the eight neighborhoods when the grid is divided, or it can be understood as the overlap of grid units. That is, for a certain geographical scene, the geographical scene corresponding to its eight neighborhoods can be regarded as its adjacent geographical scene, and the geographical scene with overlapping areas can also be regarded as its adjacent geographical scene.

[0088] Step 3.26, determine whether there is an adjacent n-1 level geographic scene node among the adjacent n-1 level geographic scene nodes of the n-level geographic scene node whose similarity with the n-level geographic scene node is greater than the similarity threshold. If so, the n-1 level geographic scene node whose similarity with the n-level geographic scene node is greater than the similarity threshold is used as the n-1 level geographic scene node subordinate to the n-level geographic scene node.

[0089] Among them, the similarity between the adjacent n-1 level geographic scene nodes and the n level geographic scene nodes in step 3.26 can also be calculated using the aforementioned similarity formula, as long as the corresponding parameters are replaced with the relevant parameters of the corresponding geographic scene nodes in step 3.26.

[0090] Step 3.27, return to step 3.24 (taking the geographical scenes corresponding to all n-1 level geographical scene nodes belonging to the n-level geographical scene node as the geographical scenes corresponding to the n-level geographical scene node) until there are no adjacent n-1 level geographical scene nodes of the n-level geographical scene node among the unvisited n-1 level geographical scene nodes, or, among the adjacent n-1 level geographical scene nodes of the n-level geographical scene node, there are no adjacent n-1 level geographical scene nodes whose similarity with the n-level geographical scene node is greater than the similarity threshold (that is, there are no n-1 level geographical scene nodes that can be merged into the n-level geographical scene node).

[0091] It should be noted that for the corresponding n Level geographic scene node n -1 level geographic scene nodes, after determining the subordinate relationship, will be in the corresponding n Level geographic scene nodes and subordinate n Level geographic scene node n -Set relationship edges between level 1 geographic scene nodes.

[0092] Exemplarily, the process of merging 0-level geographic scene nodes to generate 1-level geographic scene nodes is as follows: traverse all 0-level geographic scene nodes in sequence, and for the currently traversed 0-level geographic scene node, query its adjacent 0-level geographic scene nodes based on whether the geometric shapes overlap, and use the similarity formula to calculate its similarity with all adjacent 0-level geographic scene nodes. If the similarity is greater than the similarity threshold, merge them to generate a 1-level geographic scene node and create a relationship edge; then repeat the above operation: calculate the similarity between the newly generated 1-level geographic scene node and all its adjacent 0-level geographic scene nodes, and determine whether to perform the merge operation; repeat multiple times until there are no adjacent 0-level geographic scene nodes that can be merged. Then continue to traverse the next 0-level geographic scene node (a node that has not been visited), and continue to repeat the above operation until all 0-level geographic scene nodes have been visited. Similar to the merging process of 1-level geographic scene nodes, merge 1-level geographic scene nodes to generate 2-level geographic scene nodes; and so on; merge Level geographic scene node generation During merging, based on the merge ratio threshold Dynamically control the number of generated geographical scene nodes at different levels to make The number of level-1 geographic scene nodes is 1, realizing multi-scale geographic scene instantiation modeling.

[0093] The geographic scene instantiation modeling method provided in this application is exemplified below with reference to specific examples.

[0094] In this example, the study area is the inner urban area of ​​the first ring of Chengdu. A total of 90,267 POI point data are collected in the inner urban area of ​​the first ring of Chengdu. The data time label is 2023. The data description examples of the geographic scene nodes are shown in Table 3, and the data description examples of the geographic element nodes are shown in Table 4. Based on these data, the geographic scene instantiation model obtained by executing the above steps 1 to 3 is as follows Figure 3 shown. Figure 3 Only some of the geographic scene nodes from level 9 to level 1 are shown. Each circle in the figure represents a geographic scene node. The text in the circle represents the name of the geographic scene node. For example, level 1 scene 90 represents level 1 geographic scene No. 90, and level 2 scene 48 represents level 2 geographic scene No. 48. The figure contains the subordinate corresponding to the previous text. For example, level 2 scene 48 contains level 1 scene 90, which means that level 1 scene 90 is subordinate to level 2 scene 48. Combined with the number of geographic scene nodes at different levels shown in Table 5, it can be seen that the highest level geographic scene node is level 9, that is, level 9 scene 0 in the figure, which can be a scene of the area within the first ring of Chengdu, level 7 scene 0 in the figure can be a residential area scene, level 4 scene 8 can be a xx commercial street scene, and level 2 scene 48 can be a xx community scene. Due to the large number of scenes, they are not listed here one by one.

[0095] Table 3 Data description examples of geographic scene nodes

[0096]

[0097] Table 4 Data description examples of geographic element nodes

[0098]

[0099] In order to further verify the effectiveness of the modeling method proposed in this application, the merged geographic scene instantiation model was compared with the typical scene in reality. The results show that the modeling results based on the method in this paper can well reflect the typical geographic scenes in reality, and the geographic scene modeling results are highly consistent with the real scene. In addition, in order to verify the stability of the method in this paper when modeling geographic scenes of different scales, this paper counted the number of geographic scene nodes of different levels (scales) generated by modeling, as shown in Table 5. It can be seen that with the increase of the level, the number of geographic scene nodes decreases step by step, which effectively avoids the redundancy of the level and the instability of the scene scale, and ensures the stability and continuity of the scale transformation of the geographic scene model. The above experimental and analysis results show that this application can automatically construct a stable multi-scale geographic scene instantiation model that meets the actual needs, and provide underlying model support for the production of a new generation of knowledge service maps oriented to human needs.

[0100] Table 5 Statistics of the number of geographical scene nodes at different levels

[0101]

[0102] In summary, this application divides the study area into multiple geographic scenes, and sets the correspondence between geographic scene nodes and geographic scenes, and the correspondence between geographic element nodes and entities based on the divided geographic scenes and entities in the study area. Then, the corresponding relationship edges are set based on the association between the geographic scene nodes and the geographic element nodes to complete the construction of the geographic scene ontology model. Finally, based on the dynamically adjustable similarity threshold, the geographic scene nodes in the geographic scene ontology model are merged multiple times to realize geographic scene instantiation modeling to meet the actual application needs related to geographic scenes.

[0103] The above is a preferred embodiment of the present application. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principles described in the present application. These improvements and modifications should also be regarded as the scope of protection of the present application.

Claims

1. A geographic scene instantiation modeling method, characterized in that: include: Step 1: divide the study area into multiple grid units, and use each grid unit as a geographical scene; Step 2, constructing a geographic scene ontology model; the geographic scene ontology model includes a plurality of geographic scene nodes and a plurality of geographic element nodes, the plurality of geographic scene nodes correspond one-to-one to the plurality of grid units, each geographic scene node represents a corresponding geographic scene, the plurality of geographic element nodes correspond one-to-one to the plurality of entities in the study area, and a relationship edge is set between each geographic scene node and the geographic element node corresponding to the entity located in the corresponding grid unit; Step 3, based on the dynamically adjusted similarity threshold, the geographic scene nodes in the geographic scene ontology model are merged multiple times, and the geographic scene ontology model after the multiple merging processes is used as the geographic scene instantiation model of the study area; All geographic scene nodes in the constructed geographic scene ontology model are level 0 geographic scene nodes. After the geographic scene nodes in the geographic scene ontology model are merged for the nth time based on the dynamically adjusted similarity threshold, the highest level geographic scene node in the geographic scene ontology model is the level n geographic scene node. , Indicates the total number of levels of the preset geographic scene nodes; The n-th merging process is performed on the geographic scene nodes in the geographic scene ontology model based on the dynamically adjusted similarity threshold, including: Step 3.1, taking any n-1 level geographic scene node in the geographic scene ontology model as the current traversal node; Step 3.2, based on the set similarity threshold, the currently traversed node is used as the starting point to traverse the unvisited nodes. n -1 level geographic scene nodes are merged to obtain the corresponding nodes of the current traversal node n The geographical scene node of the level, and determine the geographical scene ontology model that is subordinate to the corresponding node of the current traversal n Level geographic scene node n -Level 1 geographic scene node; unvisited n -Level 1 geographic scene nodes are not subordinate to any n Level geographic scene node n -Level 1 geographic scene node; Step 3.3, selecting any n-1-level geographic scene node from the n-1-level geographic scene nodes that have not been visited as a new current traversal node, and returning the new current traversal node as the current traversal node in step 3.2 to execute step 3.2 until all n-1-level geographic scene nodes in the geographic scene ontology model have subordinate n-level geographic scene nodes; Step 3.4, determine whether the ratio of the number of n-level geographic scene nodes to the number of n-1-level geographic scene nodes in the geographic scene ontology model is greater than the merge ratio threshold. If so, lower the similarity threshold, delete all n-level geographic scene nodes, and return to execute step 3.1 until the ratio of the number of n-level geographic scene nodes to the number of n-1-level geographic scene nodes in the geographic scene ontology model is less than or equal to the merge ratio threshold.

2. The geographic scene instantiation modeling method according to claim 1, characterized in that: The calculation formula of the merge ratio threshold is: in, Indicates the merge ratio threshold, represents the proportionality coefficient, represents the magnification factor, .

3. The geographic scene instantiation modeling method according to claim 1, characterized in that: The step 3.2 comprises: Based on the geographical scene corresponding to the current traversal node, determine the adjacent n-1-level geographical scene node of the current traversal node from the n-1-level geographical scene nodes that have not been visited; the geographical scene corresponding to the adjacent n-1-level geographical scene node of the current traversal node is adjacent to the geographical scene corresponding to the current traversal node; Calculate the current traversal node and each adjacent n -1 level geographic scene node similarity, and the adjacent nodes whose similarity with the currently traversed node is greater than the similarity threshold n -1 level geographic scene node and the current traversal node are merged to generate the corresponding n Level geographic scene node; The current traversal node and its adjacent nodes whose similarity to the current traversal node is greater than the similarity threshold are n -1 level geographic scene node as the corresponding node of the current traversal node n Level geographic scene node n -Level 1 geographic scene node; The geographical scenes corresponding to all n-1-level geographical scene nodes belonging to the n-level geographical scene node are used as the geographical scenes corresponding to the n-level geographical scene node, and based on the geographical scenes corresponding to the n-level geographical scene node, the adjacent n-1-level geographical scene nodes of the n-level geographical scene node are determined from the n-1-level geographical scene nodes that have not been visited; the geographical scenes corresponding to the determined adjacent n-1-level geographical scene nodes are adjacent to the geographical scenes corresponding to the n-level geographical scene node; Determine whether there is an adjacent n-1 level geographic scene node whose similarity with the n-level geographic scene node is greater than the similarity threshold among the adjacent n-1 level geographic scene nodes of the n-level geographic scene node; if so, use the n-1 level geographic scene node whose similarity with the n-level geographic scene node is greater than the similarity threshold as the n-1 level geographic scene node subordinate to the n-level geographic scene node; The step of taking the geographical scenes corresponding to all n-1 level geographical scene nodes belonging to the n-level geographical scene node as the geographical scene corresponding to the n-level geographical scene node is returned until there are no adjacent n-1 level geographical scene nodes of the n-level geographical scene node among the unvisited n-1 level geographical scene nodes, or there are no adjacent n-1 level geographical scene nodes of the n-level geographical scene node whose similarity with the n-level geographical scene node is greater than the similarity threshold.

4. The geographic scene instantiation modeling method according to claim 3 is characterized in that: The calculating the similarity between the current traversal node and each adjacent n-1 level geographic scene node includes: The similarity between the current traversal node and each adjacent n-1 level geographic scene node is calculated by a similarity formula; the similarity formula is: ； in, Indicates the current traversal node The similarity with the jth adjacent n-1 level geographic scene node, represents adjustable parameters, represents the discount factor, Indicates the execution round of the merge operation. represents the area of ​​the geographic scene corresponding to the jth adjacent n-1 level geographic scene node, represents the sum of the areas of all geographic scenes, Indicates the number of entities contained in the geographic scene corresponding to the jth adjacent n-1 level geographic scene node, Represents the sum of the number of entities contained in all geographic scenes, Indicates the current traversal node The feature vector of the corresponding geographic scene, Represents the feature vector of the geographic scene corresponding to the j-th adjacent n-1-level geographic scene node.

5. The geographic scene instantiation modeling method according to claim 3, characterized in that: The geographic scene instantiation modeling method also includes: In the currently traversed node corresponding to n Level geographic scene nodes and subordinate to the n Level geographic scene node n -Set relationship edges between level 1 geographic scene nodes.

6. The geographic scene instantiation modeling method according to claim 1, characterized in that: The multiple entities include buildings, roads, and POI points.

Images