Path reconstruction methods, apparatus, computer equipment, storage media and products

Abstract

This application discloses a path reconstruction method, apparatus, computer equipment, storage medium, and product. The method involves acquiring sampled traffic paths within a region to be reconstructed, including originating and ending traffic areas; performing path clustering on the sampled traffic paths based on the originating and ending traffic areas to obtain target traffic paths; acquiring checkpoint traffic flow data obtained from traffic flow detection of target road segments within the region to be reconstructed; performing maximum entropy traffic allocation on the checkpoint traffic flow based on the target traffic paths to obtain the path flow of the target traffic paths relative to the target road segments; and performing path reconstruction on the region to be reconstructed based on the target traffic paths and path flow to obtain the path distribution and path flow within the region to be reconstructed. This application performs path reconstruction based on sampled target traffic paths and the full traffic flow detected at checkpoints, ensuring that the path reconstruction results match real-world traffic flow conditions and are more accurate.

Description

Technical Field

[0001] This application relates to the field of communication technology, specifically to a path reconstruction method, apparatus, computer equipment, storage medium, and product, wherein the storage medium is a computer-readable storage medium, and the product is a computer program product. Background Technology

[0002] Traffic flow is a fundamental component of urban traffic operations. By reconstructing traffic paths and their flow information, basic data support can be provided for applications such as traffic situation indicators, traffic simulation, and traffic control optimization. However, conventional methods for reconstructing traffic paths do not consider the actual traffic paths chosen by vehicles. They simply determine the traffic path between the starting point and the destination based on the shortest path between two points and allocate flow according to pre-set road capacity. This results in a significant difference between the obtained traffic path and the actual traffic path, leading to inaccurate path reconstruction. Summary of the Invention

[0003] This application provides a path reconstruction method, apparatus, computer device, storage medium, and product to improve the accuracy of path reconstruction results.

[0004] This application provides a path reconstruction method, including: Obtain sampled traffic paths within the area to be reconstructed. Each sampled traffic path consists of at least one road segment and includes a starting traffic area and an ending traffic area. Based on the starting traffic area and the ending traffic area, the sampled traffic path is subjected to path clustering processing to obtain the target traffic path between the starting traffic area and the ending traffic area; Obtain the traffic flow at the checkpoints by performing traffic flow detection on the target road segment within the area to be reconstructed; Based on the target traffic path, the traffic flow at the checkpoint is allocated using maximum entropy to obtain the path flow of the target traffic path with respect to the target road segment; Based on the target traffic path and the path flow, the path to be reconstructed area is reconstructed to obtain the path distribution and path flow within the area to be reconstructed.

[0005] Accordingly, embodiments of this application also provide a path reconstruction apparatus, comprising: The path acquisition unit is used to acquire sampled traffic paths within the area to be reconstructed. The sampled traffic paths consist of at least one road segment and include a starting traffic area and an ending traffic area. A clustering unit is used to perform path clustering processing on the sampled traffic path based on the starting traffic area and the ending traffic area to obtain the target traffic path between the starting traffic area and the ending traffic area; The traffic flow acquisition unit is used to acquire the traffic flow at the checkpoint obtained by traffic flow detection of the target road segment in the area to be reconstructed. The allocation unit is used to perform maximum entropy flow allocation on the traffic flow at the checkpoint based on the target traffic path, so as to obtain the path flow of the target traffic path with respect to the target road segment; The reconstruction unit is used to reconstruct the path in the area to be reconstructed based on the target traffic path and the path flow, so as to obtain the path distribution and path flow in the area to be reconstructed.

[0006] In one embodiment, the allocation unit includes: The calculation subunit is used to calculate the adjustment parameters for the path flow based on the path flow of the target traffic path with respect to at least one target road segment it contains; An adjustment subunit is used to adjust the path flow based on the adjustment parameters to obtain the initial adjusted path flow. The traffic flow allocation subunit is used to perform maximum entropy traffic flow allocation on the checkpoint traffic flow based on the initial adjusted path flow and the target traffic path, so as to obtain the adjusted path flow of the target traffic path with respect to the target road segment. The loop subunit is used to take the adjusted path flow as the path flow and return to execute the step of calculating the adjustment parameters of the path flow based on the path flow of the target traffic path with respect to at least one target road segment, until the adjustment parameters meet the preset conditions.

[0007] In one embodiment, the computing subunit includes: The selection module is used to determine a first target road segment and a second target road segment from multiple target road segments, wherein the first target road segment and the second target road segment are adjacent road segments; A parameter calculation module is used to calculate the adjustment parameter based on the ratio between the path flow of the target traffic path with respect to the first target road segment and the path flow with respect to the second target road segment. In one embodiment, the clustering unit includes: The similarity calculation subunit is used to calculate the path similarity between each sampled traffic path and other sampled traffic paths in each pair of origin traffic areas and destination traffic areas. The path clustering subunit is used to perform path clustering processing on multiple sampled traffic paths between each pair of starting traffic areas and ending traffic areas based on the path similarity, and obtain the clustering result. The target path determination subunit is used to determine the target traffic path based on the clustering results.

[0008] In one embodiment, the path clustering subunit includes: The feature information acquisition module is used to obtain the path feature information of each sampled traffic path based on the path similarity between each sampled traffic path between each pair of origin traffic areas and destination traffic areas and other sampled traffic paths between each pair of origin traffic areas and destination traffic areas. The clustering processing module is used to perform path clustering processing on the multiple sampled traffic paths based on the path feature information to obtain clustering results.

[0009] In one embodiment, the clustering result includes multiple clusters, and the target path determination subunit includes: The quantity determination module is used to determine the number of sampled traffic paths contained in each cluster based on the clustering results; The threshold determination module is used to determine the filtering threshold based on the number of paths; The filtering module is used to filter the target traffic path from the sampled traffic paths according to the filtering threshold.

[0010] In one embodiment, the path acquisition unit includes: The candidate path acquisition subunit is used to acquire candidate sampled traffic paths within the area to be reconstructed, wherein the candidate sampled traffic paths include candidate origin traffic areas and candidate destination traffic areas; The calculation subunit is used to calculate, based on the candidate sampled traffic path, the first number of times the candidate starting point traffic area is set as the starting point and the second number of times the candidate ending point traffic area is set as the ending point. A selection subunit is configured to select a starting traffic area from the candidate starting traffic areas based on the first number of times, and to select a destination traffic area from the candidate destination traffic areas based on the second number of times. The route filtering subunit is used to filter the sampled traffic routes from the candidate sampled traffic routes based on the origin traffic area and the destination traffic area.

[0011] Accordingly, this application also provides a computer device including a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to execute any of the path reconstruction methods provided in this application.

[0012] Accordingly, embodiments of this application also provide a computer-readable storage medium for storing a computer program, which is loaded by a processor to execute any of the path reconstruction methods provided in embodiments of this application.

[0013] Accordingly, this application also provides a computer program product, including a computer program, which, when executed by a processor, implements any of the path reconstruction methods provided in this application.

[0014] This application embodiment obtains sampled traffic paths within the area to be reconstructed. Each sampled traffic path consists of at least one road segment and includes a starting traffic area and an ending traffic area. Based on the starting and ending traffic areas, path clustering is performed on the sampled traffic paths to obtain the target traffic path between the starting and ending traffic areas. Traffic flow at checkpoints is obtained by traffic detection of the target road segments within the area to be reconstructed. Based on the target traffic path, maximum entropy traffic flow allocation is performed on the checkpoint traffic flow to obtain the path flow of the target traffic path with respect to the target road segments. Based on the target traffic path and path flow, path reconstruction is performed on the area to be reconstructed to obtain the path distribution and path flow within the area to be reconstructed.

[0015] This application embodiment performs path clustering on sampled traffic paths to obtain popular target traffic paths in the area to be reconstructed. Then, it reconstructs the target traffic paths based on the traffic flow at the checkpoints set up along the target traffic paths. Based on the sampled target traffic paths and the total traffic flow obtained through checkpoint detection, it expands the sampling to obtain the path distribution and path flow in the area to be reconstructed, making the path reconstruction results consistent with the actual traffic flow in the area to be reconstructed, and the path reconstruction results are more accurate. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 This is a scenario diagram of the path reconstruction method provided in the embodiments of this application; Figure 2 This is a flowchart of the path reconstruction method provided in the embodiments of this application; Figure 3 This is another flowchart of the path reconstruction method provided in the embodiments of this application; Figure 4 This is a schematic diagram of the path reconstruction device provided in an embodiment of this application; Figure 5 This is a schematic diagram of the structure of the computer device provided in the embodiments of this application. Detailed Implementation

[0018] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] This application provides a path reconstruction method, apparatus, computer device, and computer-readable storage medium. The path reconstruction apparatus can be integrated into a computer device, which may be a server or a terminal, etc.

[0020] The terminal may include mobile phones, wearable smart devices, tablets, laptops, personal computers (PCs), and in-vehicle computers, etc.

[0021] The server can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0022] For example, such as Figure 1 As shown, the computer equipment acquires sampled traffic paths within the area to be reconstructed. Each sampled traffic path consists of at least one road segment. The traffic area where the starting point of the sampled traffic path is located is the starting traffic area, and the traffic area where the ending point is located is the ending traffic area. Based on the starting and ending traffic areas, the sampled traffic paths are clustered to obtain the target traffic path between the starting and ending traffic areas. The traffic flow at the checkpoints within the area to be reconstructed is obtained through traffic flow detection of the target road segments. Based on the target traffic path, the maximum entropy traffic flow allocation is performed on the checkpoint traffic flow to obtain the path flow of the target traffic path with respect to the target road segments. Based on the target traffic path and path flow, the path is reconstructed within the area to be reconstructed to obtain the path distribution and path flow within the area to be reconstructed.

[0023] The embodiments of this application are based on the traffic paths of the area to be reconstructed obtained by sampling, i.e. the target traffic paths, and the total traffic flow obtained by checkpoint detection, i.e. the checkpoint traffic flow. The path distribution and path flow of the area to be reconstructed are expanded to make the path reconstruction results consistent with the actual traffic flow of the area to be reconstructed, and the path reconstruction results are more accurate.

[0024] The following sections provide detailed descriptions of each example. It should be noted that the order in which the embodiments are described is not intended to limit the preferred order of the embodiments.

[0025] This embodiment will be described from the perspective of a path reconstruction device, which can be integrated into a computer device, such as a server or a terminal.

[0026] This application provides a path reconstruction method, such as... Figure 2 As shown, the specific process of this path reconstruction method can be as follows: 101. Obtain the sampled traffic paths within the area to be reconstructed. Each sampled traffic path consists of at least one road segment and includes a starting traffic area and an ending traffic area.

[0027] In the field of transportation, a road segment is a traffic route between two adjacent nodes on a transportation network. A series of connected road segments arranged in an orderly fashion constitutes a path. Therefore, a sampled traffic path can consist of at least one road segment, including a starting point and an ending point. The traffic area where the starting point is located is the starting point traffic area, and the traffic area where the ending point is located is the ending point traffic area. A traffic area can be a closed polygonal region enclosed by roads of a certain level. Specifically, it can be a traffic zone in the field of transportation planning and management, often used as a basic spatial unit for research and analysis in traffic management, urban planning, and urban governance. Alternatively, a traffic area can be a region defined by technical personnel. For example, a square area with a length of 5 kilometers and a width of 5 kilometers can be considered a traffic area. There are various ways to divide traffic areas, and this invention does not limit this; technical personnel can divide the area according to actual application needs.

[0028] Sampling traffic routes can be obtained through cameras, and these routes can be travel paths, such as vehicle routes (also known as vehicle trajectories) or pedestrian walking routes. Alternatively, they can be obtained by using the user's geographic location collected by the terminal when providing navigation services to the user through a client application.

[0029] The sampled traffic paths in the area to be reconstructed can be paths between popular origin traffic areas and popular destination traffic areas. Sampled traffic paths can be filtered based on the frequency of occurrence of the origin and destination traffic areas. In one embodiment, the step "obtaining multiple sampled traffic paths within the area to be reconstructed" can specifically include: Obtain candidate sampled traffic routes within the area to be reconstructed. The candidate sampled traffic routes include candidate origin traffic areas and candidate destination traffic areas. Based on the candidate sampled traffic paths, calculate the first number of times the candidate starting point traffic area is set as the starting point and the second number of times the candidate ending point traffic area is set as the ending point. The starting traffic area is selected from the candidate starting traffic areas based on the first count, and the ending traffic area is selected from the candidate ending traffic areas based on the second count; Based on the origin and destination traffic areas, sampled traffic routes are selected from the candidate sampled traffic routes.

[0030] The candidate sampling traffic routes can be travel routes within the area to be reconstructed, captured by cameras, such as vehicle travel paths (also known as vehicle trajectories or driving trajectories) or pedestrian walking routes. Alternatively, they can be obtained by using the user's geographical location collected by the terminal when providing navigation services to the user through a client application.

[0031] The candidate traffic route includes a candidate origin traffic area and a candidate destination traffic area. The candidate origin traffic area is the starting point of the candidate traffic route, and the candidate destination traffic area is the destination of the candidate traffic route.

[0032] The number of candidate origin and destination traffic areas in multiple candidate traffic routes is counted to obtain the number of times each candidate origin traffic area is selected as the departure point (i.e., the first count) and the number of times each candidate destination traffic area is selected as the destination (i.e., the second count). Candidate origin traffic areas with a first count greater than a first threshold are selected to obtain the origin traffic area, and candidate destination traffic areas with a second count greater than a second threshold are selected to obtain the destination traffic area. The origin traffic area is the popular traffic area that users choose as the departure point, and the destination traffic area is the popular traffic area that users choose as the destination.

[0033] Candidate traffic routes that start from the originating traffic area and end at the destination traffic area are selected as sampled traffic routes.

[0034] 102. Based on the origin traffic area and the destination traffic area, perform path clustering on the sampled traffic paths to obtain the target traffic paths between the origin traffic area and the destination traffic area.

[0035] For example, for each pair of origin and destination traffic areas, path clustering is performed based on the similarity of their corresponding multiple sampled traffic paths to select popular target traffic paths between the pair of origin and destination traffic areas from the multiple sampled traffic paths. In one embodiment, the step "performing path clustering based on origin and destination traffic areas to obtain target traffic paths between origin and destination traffic areas" may specifically include: Calculate the path similarity between each sampled traffic path among multiple sampled traffic paths between each pair of origin traffic areas and destination traffic areas, and the other sampled traffic paths between each pair of origin traffic areas and destination traffic areas; Based on path similarity, multiple sampled traffic paths between each pair of origin and destination traffic areas are clustered to obtain clustering results. The target traffic route is determined based on the clustering results.

[0036] Each pair of origin and destination traffic areas refers to the traffic area at the start and end of a path. There may be one, two, or more paths between the same pair of origin and destination traffic areas. This step is to discover popular paths between a pair of origin and destination traffic areas.

[0037] Among them, path similarity can be used to measure the degree of similarity between two sampled traffic paths. The higher the similarity, the smaller the difference between the two sampled traffic paths.

[0038] For example, specifically, path features can be extracted for each sampled traffic path to obtain the path feature information corresponding to each sampled traffic path. The path feature information is the information that represents the sampled traffic path in the feature space, and the path feature information can be in the form of a feature matrix, etc.

[0039] For each pair of origin and destination traffic areas, the path similarity between any two sampled traffic paths is calculated based on the distance between the path feature information of each sampled traffic path. Similar sampled traffic paths are then grouped together based on their path similarity to perform path clustering processing. The clustering results can include at least one cluster, which contains sampled traffic paths and cluster centers. Based on the number of sampled traffic paths contained in the cluster, the cluster with the most samples can be selected, and the sampled traffic path corresponding to the cluster center of the cluster is taken as the target traffic path. In other words, the frequently traveled traffic paths between the origin and destination traffic areas are extracted from multiple sampled traffic paths.

[0040] The distance can be cosine distance, Manhattan distance, Hamming distance, or Euclidean distance, etc.

[0041] Optionally, path feature information of each sampled traffic path can be obtained based on the similarity between each sampled traffic path and other sampled traffic paths, and then clustering processing can be performed. That is, in one embodiment, the step "perform path clustering processing on multiple sampled traffic paths between each pair of origin traffic areas and destination traffic areas based on path similarity to obtain clustering results" can specifically include: Based on the path similarity between each sampled traffic path between each pair of origin traffic areas and destination traffic areas and other sampled traffic paths between each pair of origin traffic areas and destination traffic areas, the path feature information of each sampled traffic path is obtained. Based on path feature information, multiple sampled traffic paths are subjected to path clustering processing to obtain clustering results.

[0042] For example, in one embodiment, path features can be extracted for each sampled traffic path to obtain initial path feature information corresponding to each sampled traffic path. The initial path feature information is information that represents the sampled traffic path in the feature space, and the initial path feature information can be in the form of a feature matrix or the like.

[0043] Then, for each pair of origin and destination traffic areas, the similarity between each sampled traffic path and other sampled traffic paths is calculated based on the initial path feature information. This similarity is used as the path feature information for that sampled traffic path. The path feature information includes the similarity to other sampled traffic paths. Clustering is performed based on the path feature information of the sampled traffic paths to obtain clustering results. The sampled traffic path corresponding to the cluster center of each cluster is used as the representative path of that cluster, and the number of sampled traffic paths contained in that cluster is used as the number of representative paths. A larger number of representative paths can be used as the target traffic paths. Optionally, the number of representative paths can be compared with a screening threshold, and representative paths exceeding the preset threshold can be used as the target traffic paths.

[0044] The filtering threshold can be a pre-set threshold for selecting target traffic routes. Optionally, the filtering threshold can also be determined based on the proportion of the selected target traffic routes among the number of sampled traffic routes, so that the selected target traffic routes can represent the majority of the sampled traffic routes. That is, in one embodiment, the clustering result includes multiple clusters. The step "determine the target traffic route based on the clustering result" can specifically include: The number of sampled traffic paths contained in each cluster is determined based on the clustering results; Determine the filtering threshold based on the number of paths; The target traffic route is obtained by filtering from the sampled traffic routes based on the filtering threshold.

[0045] For example, it can be done by pre-setting the proportion of the selected target traffic path in the sampled traffic paths (e.g., 95%), determining the number of sampled traffic paths contained in each cluster in the clustering results, sorting the number of sampled traffic paths contained in the clusters, determining the clusters to be selected based on the pre-set proportion, then determining the screening threshold, screening the clusters containing more than the screening threshold, and using the sampled traffic path corresponding to the cluster center as the target traffic path.

[0046] 103. Obtain the traffic flow at checkpoints obtained by traffic flow detection of the target road segment within the area to be reconstructed.

[0047] Traffic flow at checkpoints can be obtained by setting up checkpoints at intersections or other locations on the target road segment and detecting the flow of people or vehicles passing through these checkpoints. Therefore, traffic flow at checkpoints can represent the flow of people passing through the corresponding checkpoints. If a checkpoint is set at an intersection on target road segment a, and that intersection is also an intersection on target road segment b, then the traffic flow at checkpoints is the traffic flow of both target road segments a and b. In other words, the traffic flow at checkpoints is the traffic flow of at least one segment of the target road segment corresponding to the set checkpoint. If a checkpoint is set in the middle of the target road segment, then the traffic flow at checkpoints is the traffic flow of the target road segment.

[0048] The traffic flow collected at the checkpoint is the total traffic flow passing through the checkpoint within a certain time period, while the sampled traffic path is a sampled part of the traffic path. In this embodiment, the sampling can be expanded based on the total traffic flow and the target traffic path to obtain the path distribution and path flow that conform to the actual situation in the area to be reconstructed.

[0049] Traffic flow at checkpoints can be detected by devices composed of geomagnetic coils or cameras. Traffic flow at checkpoints can include pedestrian flow, vehicle flow, mobile device flow, etc., without limitation.

[0050] 104. Based on the target traffic path, perform maximum entropy flow allocation on the traffic flow at the checkpoint to obtain the path flow of the target traffic path with respect to the target road segment.

[0051] Among them, maximum entropy traffic allocation can be based on the principle of maximum entropy to allocate traffic flow at checkpoints, in order to determine the proportion of traffic flow of each target traffic path passing through the checkpoint in the traffic flow of the target road segment.

[0052] Since a checkpoint can correspond to multiple target road segments, and these multiple target road segments belong to different target traffic paths, the path traffic flow of a target traffic path cannot be determined based on the checkpoint traffic flow corresponding to each checkpoint included in that target traffic path. The path traffic flow of each target traffic path can only be determined based on the number of all target traffic paths passing through that checkpoint, as well as the checkpoint traffic flow corresponding to other target road segments within that target traffic path.

[0053] For example, for a target traffic route with multiple checkpoints, since vehicles traveling on the target traffic route do not disappear out of thin air, the traffic flow of the target traffic route with respect to the multiple checkpoints should be the same. Suppose that there are checkpoints a and b on the target traffic route a, then the traffic flow of the target traffic route with respect to checkpoint a and checkpoint b should be the same. Therefore, the traffic flow of different checkpoints is mutually constrained, and the traffic flow distribution of other checkpoints can be constrained based on the traffic flow of one checkpoint on the target traffic route.

[0054] Suppose that there are road segments a, b, c and d in the area to be reconstructed. Road segments a, b, c and d are road segments that make up two target traffic paths. Target traffic path A includes road segments a, c and d, and target traffic path B includes road segments a, b and c.

[0055] Checkpoint a is set up on road segment a, and the traffic flow detected through checkpoint a is 20; checkpoint b is set up on road segment b, and the traffic flow detected through checkpoint b is 10; checkpoint c is set up on road segment c, and the traffic flow detected through checkpoint c is 20; checkpoint d is set up on road segment d, and the traffic flow detected through checkpoint d is 10.

[0056] Since only target traffic path A includes road segment d, and only target traffic path B includes road segment b, the allocation of traffic flow to road segments A and B at other checkpoints can be constrained based on the checkpoint traffic flow of road segment d and road segment b. The allocation of traffic flow at other checkpoints will also adjust the traffic flow allocated from checkpoint b to target traffic path B and from checkpoint d to target traffic path A. This allows for traffic flow allocation to target traffic paths based on checkpoint traffic flow, resulting in the path flow of each target traffic path. Based on the path flow corresponding to each target traffic path, the path distribution and path flow within the area to be reconstructed can be obtained.

[0057] In one embodiment, the target traffic path and the corresponding checkpoint traffic flow can be input into a neural network model. The neural network model traverses each checkpoint and allocates the checkpoint traffic flow to the target traffic path passing through the checkpoint based on the maximum entropy principle.

[0058] The neural network model can be iterated continuously based on the constraints between the path flow of different checkpoints on the target traffic path until the path flow of each traffic path with respect to each checkpoint is equivalent, thus obtaining the path flow of each traffic path with respect to the target road segment, that is, the component of the checkpoint flow from each target traffic path.

[0059] For example, the traffic flow at the checkpoint can be arbitrarily taken as the actual value. The predicted path flow of each target traffic path can be predicted by a neural network model. The traffic flow calculation amount at the checkpoint (referred to as the calculation amount) can be calculated based on the predicted path flow of each target traffic path. Based on the difference between the calculation amount and the actual value, backpropagation is performed to adjust the model parameters of the neural network model. The predicted path flow of the target traffic path is then predicted. The above process is iterated until the difference between the calculation amount and the actual value meets the preset condition, such as being less than a preset threshold, and the path flow of each target traffic path is obtained.

[0060] In one embodiment, the maximum entropy traffic flow allocation for the checkpoint can specifically be as follows: assuming a checkpoint E is set up in the area to be reconstructed, and the checkpoint traffic flow is... There are K target traffic routes passing through checkpoint E. The initial path flow of the target traffic route is Q. Based on the maximum entropy principle, the probability P that the path flow of the target traffic path is equal to the initial path flow Q is calculated. The maximum entropy principle involves selecting the model with the maximum entropy from a set of known constraints. The constraints for allocating traffic flow at checkpoints based on the maximum entropy principle can be... Therefore, based on the principle of maximum entropy, the probability corresponding to each target traffic path can be obtained. Based on the probability, the initial path flow can be adjusted to obtain the path flow. The initial path flow can be preset; for example, the initial path flow can be preset to 1.

[0061] Optionally, for a checkpoint, the initial path flow at that checkpoint can be the path flow allocated to the target traffic path from the path flow of the previous checkpoint that arrives earlier than that checkpoint in the target traffic path. Then, based on the path flow at that checkpoint, the path flow of the previous checkpoint is adjusted according to the path flow at that checkpoint and the path flow of the previous checkpoint. That is, in one embodiment, each target traffic path contains at least one target road segment. The step "based on the target traffic path, perform maximum entropy flow allocation on the checkpoint traffic flow to obtain the path flow of the target traffic path with respect to the target road segment" can specifically include: Calculate the adjustment parameters for the path flow based on the path flow of the target traffic path with respect to at least one target road segment it contains; The path flow is adjusted based on the adjustment parameters to obtain the initial adjusted path flow. Based on the initial adjusted path flow and the target traffic path, the maximum entropy flow allocation is performed on the checkpoint traffic flow to obtain the adjusted path flow of the target traffic path with respect to the target road segment. The adjusted path flow is used as the path flow. The process returns to calculate the adjustment parameters for the path flow based on the path flow of the target traffic path with respect to at least one target road segment, until the adjustment parameters meet the preset conditions.

[0062] Among them, the adjustment parameters can be parameters used to adjust the path traffic.

[0063] For example, it could be a target traffic route with multiple checkpoints. Since the starting and ending traffic areas of the target traffic route are fixed, the target traffic route is directional. The arrival order of the multiple checkpoints is different, and they can be sorted according to the arrival order. The first checkpoint to arrive is the first checkpoint, the second checkpoint to arrive is the second checkpoint, and so on.

[0064] Referring to the above constraints Methods for calculating path flow, calculating the target traffic path Path traffic at the first checkpoint M and will The initial path flow for the second checkpoint N Then, based on the initial path flow of the second checkpoint and the constraints of the second checkpoint, the path flow of the second checkpoint is calculated. It can be understood that the path flow of target traffic paths that do not pass through the second checkpoint is zero.

[0065] Then, calculate the ratio of the path traffic at the second checkpoint to the path traffic at the first checkpoint, and determine the adjustment parameters based on this ratio. For example, it can be based on... Calculating adjustment parameters using an idempotent function Calculate the path flow after initial adjustment based on the adjustment parameters. .

[0066] Based on the above method, the initial adjusted path flow of each target traffic path in the area to be reconstructed with respect to the first checkpoint can be calculated. Then, based on the initial adjusted path flow and constraints... Calculated based on the maximum entropy principle Target traffic route The adjusted path flow rate for the first checkpoint is: .

[0067] target traffic route The adjusted path flow rate for the first checkpoint is: Target traffic route The path traffic at the first checkpoint is used as the basis for the path traffic at the second checkpoint. The path traffic at the first checkpoint is then adjusted based on the path traffic at the second checkpoint. This adjustment continues until the ratio of the path traffic at the second checkpoint to the path traffic at the first checkpoint is less than a preset threshold.

[0068] Then, the flow rate of the second checkpoint N is adjusted. The adjustment process can be referred to the process of adjusting the flow rate of the first checkpoint M, which will not be elaborated here.

[0069] It is understandable that once the traffic flow at the second checkpoint is adjusted, the traffic flow at the first checkpoint can be adjusted based on that traffic flow.

[0070] By continuously adjusting the traffic flow at checkpoints along the target traffic route, the traffic flow along the target traffic route is made to match the actual traffic flow.

[0071] Optionally, the adjustment parameter can also be determined based on the ratio between the path flow of two adjacent target road segments, i.e., the step "calculate the adjustment parameter for path flow based on the path flow of the target traffic path with respect to at least one target road segment it contains", which may specifically include: From multiple target road segments, a first target road segment and a second target road segment are determined, and the first target road segment and the second target road segment are adjacent road segments; The adjustment parameters are calculated based on the ratio between the path flow of the target traffic path with respect to the first target road segment and the path flow with respect to the second target road segment.

[0072] For example, from multiple target road segments included in the target traffic path, two adjacent target road segments are identified: the first target road segment and the second target road segment. Based on the path flow of the first target road segment calculated using the maximum entropy principle, and the path flow of the second target road segment, the path flow ratio of the first target road segment and the second target road segment is calculated. Then, adjustment parameters are determined based on this path flow ratio. For example, it can be based on... Calculate and adjust parameters using an idempotent function.

[0073] 105. Based on the target traffic path and path flow, perform path reconstruction on the area to be reconstructed to obtain the path distribution and path flow within the area to be reconstructed.

[0074] Among them, path distribution can represent the travel paths in the area to be reconstructed, and path flow can represent the flow of the paths (or road segments) in the area to be reconstructed. Since there are intersecting road segments between the target traffic paths, the path flow in the area to be reconstructed can be obtained based on the path flow of each target traffic path.

[0075] For example, the main traffic routes in the area to be reconstructed can be determined based on the target traffic routes, and the traffic flow of each target traffic route can be determined based on the route flow. Then, based on the sampled traffic routes, the sampled traffic flow can be expanded with the full amount of traffic flow collected through the checkpoints to obtain the route distribution and route flow in the area to be reconstructed.

[0076] As can be seen from the above, the embodiments of this application obtain sampled traffic paths within the area to be reconstructed. Each sampled traffic path consists of at least one road segment and includes a starting traffic area and an ending traffic area. Based on the starting and ending traffic areas, path clustering is performed on the sampled traffic paths to obtain the target traffic path between the starting and ending traffic areas. The checkpoint traffic flow is obtained by traffic flow detection of the target road segment within the area to be reconstructed. Based on the target traffic path, maximum entropy traffic flow allocation is performed on the checkpoint traffic flow to obtain the path flow of the target traffic path with respect to the target road segment. Based on the target traffic path and path flow, path reconstruction is performed on the area to be reconstructed to obtain the path distribution and path flow within the area to be reconstructed.

[0077] This application embodiment performs path clustering on sampled traffic paths to obtain popular target traffic paths in the area to be reconstructed. Then, it reconstructs the target traffic paths based on the traffic flow at the checkpoints set up along the target traffic paths. Based on the sampled target traffic paths and the total traffic flow obtained through checkpoint detection, it expands the sampling to obtain the path distribution and path flow in the area to be reconstructed, making the path reconstruction results consistent with the actual traffic flow in the area to be reconstructed, and the path reconstruction results are more accurate.

[0078] Based on the above embodiments, the following examples will provide further detailed explanations.

[0079] This embodiment will take the traffic path as the vehicle's driving trajectory as an example and describe it from the perspective of the path reconstruction device. The path reconstruction device can be integrated into a computer device, which can be a server or a terminal or other devices. This application provides a path reconstruction method, such as... Figure 3 As shown, the specific process of this path reconstruction method can be as follows: 201. The server obtains candidate sampled traffic paths within the area to be reconstructed, and mines the target starting traffic area and the target ending traffic area from the candidate sampled traffic paths.

[0080] The target origin traffic area is the origin traffic area in the above embodiments, and the target terminal traffic area is the destination traffic area in the above embodiments.

[0081] For example, candidate sampled traffic paths within the area to be reconstructed can be collected through cameras, such as vehicle travel paths (also known as vehicle trajectories); alternatively, when providing navigation services to users through a client on the terminal, the user's travel path (i.e., driving trajectory) can be obtained by collecting the user's geographical location from the terminal, thus obtaining candidate sampled traffic paths within the area to be reconstructed.

[0082] The candidate traffic route includes a candidate origin traffic area and a candidate destination traffic area. The candidate origin traffic area is the starting point of the candidate traffic route, and the candidate destination traffic area is the destination of the candidate traffic route.

[0083] The server counts the number of candidate origin and destination traffic areas in multiple candidate traffic routes, obtaining the number of times each candidate origin traffic area is selected as the departure point (i.e., the first count) and the number of times each candidate destination traffic area is selected as the destination (i.e., the second count). The server selects the candidate origin traffic areas whose first count is greater than a first threshold to obtain the target origin traffic area, and selects the candidate destination traffic areas whose second count is greater than a second threshold to obtain the target destination traffic area. The origin traffic area is the popular traffic area that the user selects as the departure point, and the destination traffic area is the popular traffic area that the user selects as the destination.

[0084] 202. The server filters the candidate sampled traffic paths corresponding to the target origin traffic area and the target destination traffic area to obtain the sampled traffic paths.

[0085] The server filters out candidate traffic routes from those with the starting point being the originating traffic area and the destination being the ending traffic area, and uses these as sampled traffic routes.

[0086] 203. The server performs path clustering on the sampled traffic paths based on the origin traffic area and the destination traffic area to obtain the target traffic path.

[0087] For example, the server can extract path features for each sampled traffic path to obtain path feature information corresponding to each sampled traffic path. The path feature information is the information that represents the sampled traffic path in the feature space, and the path feature information can be in the form of a feature matrix, etc.

[0088] For each pair of origin and destination traffic areas, the path similarity between any two sampled traffic paths is calculated based on the distance between the path feature information of each sampled traffic path. Similar sampled traffic paths are then grouped together based on their path similarity to perform path clustering processing. The clustering results can include at least one cluster, which contains sampled traffic paths and cluster centers. Based on the number of sampled traffic paths contained in the cluster, the cluster with the most samples can be selected, and the sampled traffic path corresponding to the cluster center of the cluster is taken as the target traffic path. In other words, the frequently traveled traffic paths between the origin and destination traffic areas are extracted from multiple sampled traffic paths.

[0089] 204. The server obtains the traffic flow at the checkpoints corresponding to the road segments included in the target traffic path.

[0090] For example, checkpoint traffic flow can be obtained by setting up checkpoints at intersections or designated road sections to detect the traffic flow passing through the intersections and designated road sections. Therefore, checkpoint traffic flow can represent the traffic flow of the corresponding road section or the road section where the corresponding intersection is located. Checkpoint traffic flow can be considered as the total traffic flow of the corresponding road section, while sampled traffic paths are only the traffic paths obtained from a portion of the sampled vehicles.

[0091] Traffic flow at checkpoints can be detected by devices composed of geomagnetic coils or cameras. Traffic flow at checkpoints can include pedestrian flow, vehicle flow, mobile device flow, etc., without limitation.

[0092] 205. The server determines the matching traffic path corresponding to the traffic flow of each checkpoint based on the traffic flow of each road segment included in the target traffic path.

[0093] For example, based on the traffic flow at each checkpoint corresponding to each road segment in the target traffic path, the server can determine the matching traffic path corresponding to the checkpoint traffic flow. The matching traffic path corresponding to the checkpoint traffic flow is the target traffic path of the road segment that has the checkpoint traffic flow detected.

[0094] 206. The server matches the traffic paths corresponding to the traffic flow at each checkpoint, and performs path analysis within the area to be reconstructed to obtain the path distribution and path flow within the area to be reconstructed.

[0095] Based on the principle of maximum entropy, the sum of traffic flows of the target paths through the checkpoints can be considered as the checkpoint traffic flow, which can be regarded as a known condition. Under the constraint of the known condition, the path reconstruction of the area to be reconstructed can be carried out based on the traffic flow of each checkpoint according to the principle of maximum entropy. The checkpoint traffic flow is allocated to the target traffic paths to obtain the path flow of each target traffic path in the area to be reconstructed, and thus the path distribution and path flow of the area to be reconstructed are obtained.

[0096] Specifically, for each checkpoint, the target traffic path passing through that checkpoint can be determined. The sum of the traffic flow of the target traffic paths passing through that checkpoint is the checkpoint traffic flow. Based on the clustering results, the number of sampled traffic paths contained in the cluster of each target traffic path and the proportion of the total number of sampled traffic paths are used to determine the ratio between the traffic flow of each target traffic path and the checkpoint traffic flow. The more sampled traffic paths contained in the cluster of the target traffic path, the greater the proportion of the traffic flow of that target traffic path in the checkpoint traffic flow.

[0097] A reference traffic flow can be preset and used as the initial traffic flow for each target traffic path. The checkpoint traffic volume is calculated based on the initial traffic flow and the proportion of traffic flow for each target traffic path. The traffic flow for each target traffic path is updated based on the ratio between the checkpoint traffic volume and the checkpoint traffic flow. This process is iterated until the ratio between the checkpoint traffic flow and the checkpoint traffic volume for each checkpoint meets the preset accuracy condition. The traffic flow of each target traffic path at this point is then used as the path flow for that target traffic path. This allows us to obtain the path distribution and path flow within the area to be reconstructed.

[0098] The above process can be implemented using a neural network model. The traffic flow at the checkpoint is considered as the actual value. The neural network model predicts the predicted path flow for each target traffic path. Based on the predicted path flow for each target traffic path, the traffic flow calculation amount at the checkpoint (referred to as the calculation amount) can be calculated. Based on the difference between the calculation amount and the actual value, backpropagation is performed to adjust the model parameters of the neural network model. Then, the predicted path flow for the target traffic path is predicted again. The above process is iterated until the difference between the calculation amount and the actual value meets the preset condition, such as being less than a preset threshold, thus obtaining the path distribution and path flow in the area to be reconstructed.

[0099] As can be seen from the above, in this embodiment, the server obtains candidate sampled traffic paths within the area to be reconstructed, and mines the target origin traffic area and target destination traffic area from the candidate sampled traffic paths; it filters the candidate sampled traffic paths corresponding to the target origin traffic area and the target destination traffic area to obtain sampled traffic paths; it performs path clustering processing on the sampled traffic paths based on the origin traffic area and the destination traffic area to obtain the target traffic path; it obtains the checkpoint traffic flow corresponding to the road segments included in the target traffic path; it determines the matching traffic path corresponding to each checkpoint traffic flow based on the checkpoint traffic flow corresponding to each road segment included in the target traffic path; and it performs path processing within the area to be reconstructed based on the matching traffic path corresponding to each checkpoint traffic flow to obtain the path distribution and path flow within the area to be reconstructed.

[0100] This application embodiment performs path clustering on sampled traffic paths to obtain popular target traffic paths in the area to be reconstructed. Then, it reconstructs the target traffic paths based on the traffic flow at the checkpoints set up along the target traffic paths. Based on the sampled target traffic paths and the total traffic flow obtained through checkpoint detection, it expands the sampling to obtain the path distribution and path flow in the area to be reconstructed, making the path reconstruction results consistent with the actual traffic flow in the area to be reconstructed, and the path reconstruction results are more accurate.

[0101] To facilitate better implementation of the path reconstruction method provided in the embodiments of this application, a path reconstruction apparatus is also provided in one embodiment. The meanings of the terms used are the same as in the path reconstruction method described above, and specific implementation details can be found in the description of the method embodiments.

[0102] This path reconstruction device can be integrated into a computer device, such as... Figure 4 As shown, the path reconstruction device may include: a path acquisition unit 301, a clustering unit 302, a traffic acquisition unit 303, an allocation unit 304, and a reconstruction unit 305, as detailed below: (1) Path acquisition unit 301: used to acquire sampled traffic paths in the area to be reconstructed. The sampled traffic path consists of at least one road segment and includes the starting traffic area and the ending traffic area.

[0103] In one embodiment, the path acquisition unit 301 may include a candidate path acquisition subunit, a calculation subunit, a selection subunit, and a path filtering subunit, specifically: Candidate Path Acquisition Subunit: Used to acquire candidate sampled traffic paths within the area to be reconstructed. Candidate sampled traffic paths include candidate origin traffic areas and candidate destination traffic areas. Calculation subunit: used to calculate the first number of times the candidate starting point traffic area is set as the starting point and the second number of times the candidate ending point traffic area is set as the ending point, based on the candidate sampled traffic paths; Selecting sub-units: used to select the starting traffic area from the candidate starting traffic areas based on the first count, and to select the ending traffic area from the candidate ending traffic areas based on the second count; Route filtering sub-unit: used to filter sampled traffic routes from candidate sampled traffic routes based on the origin traffic area and the destination traffic area.

[0104] (2) Clustering unit 302: is used to perform path clustering processing on the sampled traffic paths based on the starting traffic area and the ending traffic area to obtain the target traffic path between the starting traffic area and the ending traffic area.

[0105] In one embodiment, the clustering unit may include a similarity calculation subunit, a path clustering subunit, and a target path determination subunit, specifically: Similarity calculation subunit: used to calculate the path similarity between each sampled traffic path and other sampled traffic paths in multiple sampled traffic paths between each pair of origin traffic areas and destination traffic areas; Path clustering subunit: Used to perform path clustering on multiple sampled traffic paths between each pair of origin traffic areas and destination traffic areas based on path similarity, and obtain the clustering results; Target route determination sub-unit: used to determine the target traffic route based on clustering results.

[0106] In one embodiment, the path clustering subunit may include a feature information acquisition module and a clustering processing module, specifically: Feature information acquisition module: used to obtain the path feature information of each sampled traffic path based on the path similarity between each sampled traffic path between each pair of origin traffic areas and destination traffic areas and other sampled traffic paths between each pair of origin traffic areas and destination traffic areas; Clustering module: Used to perform path clustering on multiple sampled traffic paths based on path feature information to obtain clustering results.

[0107] In one embodiment, the clustering result includes multiple clusters, and the target path determination subunit may include a quantity determination module, a threshold determination module, and a filtering module, specifically: Quantity determination module: used to determine the number of sampled traffic paths contained in each cluster based on the clustering results; Threshold determination module: used to determine the filtering threshold based on the number of paths; Filtering module: Used to filter the target traffic routes from the sampled traffic routes based on the filtering threshold.

[0108] (3) Traffic acquisition unit 303: used to acquire the traffic flow at the checkpoint obtained by traffic detection of the target road segment in the area to be reconstructed.

[0109] (4) Allocation unit 304: used to reconstruct the path in the area to be reconstructed based on the target traffic path and path flow, so as to obtain the path distribution and path flow in the area to be reconstructed.

[0110] In one embodiment, the allocation unit 304 may include a calculation subunit, an adjustment subunit, a flow allocation subunit, and a circulation subunit, specifically: Calculation subunit: used to calculate the adjustment parameters for path flow based on the path flow of the target traffic path with respect to at least one target road segment it contains; Adjustment subunit: used to adjust the path flow based on adjustment parameters to obtain the initial adjusted path flow; Traffic flow allocation subunit: Used to perform maximum entropy traffic flow allocation on checkpoint traffic flow based on the initial adjusted path flow and the target traffic path, so as to obtain the adjusted path flow of the target traffic path with respect to the target road segment; Loop subunit: Used to take the adjusted path flow as the path flow and return to execute the calculation of the adjustment parameters for the path flow based on the path flow of the target traffic path with respect to at least one target road segment, until the adjustment parameters meet the preset conditions.

[0111] In one embodiment, the calculation subunit may include a selection module and a parameter calculation module, specifically: Selection module: used to determine the first target road segment and the second target road segment from multiple target road segments, wherein the first target road segment and the second target road segment are adjacent road segments; Parameter calculation module: used to calculate adjustment parameters based on the ratio between the path flow of the target traffic path with respect to the first target road segment and the path flow with respect to the second target road segment.

[0112] (5) Reconstruction unit 305: used to reconstruct the path in the area to be reconstructed based on the target traffic path and path flow, to obtain the path distribution and path flow in the area to be reconstructed. As can be seen from the above, the path reconstruction device in this application embodiment obtains the sampled traffic path in the area to be reconstructed through the path acquisition unit 301. The sampled traffic path consists of at least one road segment and includes a starting traffic area and an ending traffic area. The clustering unit 302 performs path clustering processing on the sampled traffic path based on the starting traffic area and the ending traffic area to obtain the target traffic path between the starting traffic area and the ending traffic area. The flow acquisition unit 303 obtains the checkpoint traffic flow obtained by flow detection of the target road segment in the area to be reconstructed. The allocation unit 304 performs maximum entropy flow allocation on the checkpoint traffic flow based on the target traffic path to obtain the path flow of the target traffic path with respect to the target road segment. The reconstruction unit 305 reconstructs the path in the area to be reconstructed based on the target traffic path and path flow, to obtain the path distribution and path flow in the area to be reconstructed.

[0113] This application embodiment performs path clustering on sampled traffic paths to obtain popular target traffic paths in the area to be reconstructed. Then, it reconstructs the target traffic paths based on the traffic flow at the checkpoints set up along the target traffic paths. Based on the sampled target traffic paths and the total traffic flow obtained through checkpoint detection, it expands the sampling to obtain the path distribution and path flow in the area to be reconstructed, making the path reconstruction results consistent with the actual traffic flow in the area to be reconstructed, and the path reconstruction results are more accurate.

[0114] This application also provides a computer device, which can be a terminal or a server, such as... Figure 5 As shown, it illustrates a structural schematic diagram of the computer device involved in the embodiments of this application, specifically: The computer device may include components such as a processor 1001 with one or more processing cores, a memory 1002 with one or more computer-readable storage media, a power supply 1003, and an input unit 1004. Those skilled in the art will understand that... Figure 5 The computer device structure shown does not constitute a limitation on the computer device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: The processor 1001 is the control center of the computer device. It connects various parts of the computer device via various interfaces and lines. By running or executing software programs and / or modules stored in the memory 1002, and by calling data stored in the memory 1002, it performs various functions of the computer device and processes data, thereby performing overall detection of the computer device. Optionally, the processor 1001 may include one or more processing cores; preferably, the processor 1001 may integrate an application processor and a modem processor, wherein the application processor mainly handles the operating system, user interface, and computer programs, and the modem processor mainly handles wireless communication. It is understood that the modem processor may not be integrated into the processor 1001.

[0115] The memory 1002 can be used to store software programs and modules. The processor 1001 executes various functional applications and data processing by running the software programs and modules stored in the memory 1002. The memory 1002 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, computer programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the computer device, etc. In addition, the memory 1002 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 1002 may also include a memory controller to provide the processor 1001 with access to the memory 1002.

[0116] The computer equipment also includes a power supply 1003 that supplies power to the various components. Preferably, the power supply 1003 can be logically connected to the processor 1001 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 1003 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0117] The computer device may also include an input unit 1004, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0118] Although not shown, the computer device may also include a display unit, etc., which will not be described in detail here. Specifically, in this embodiment, the processor 1001 in the computer device loads the executable files corresponding to the processes of one or more computer programs into the memory 1002 according to the following instructions, and the processor 1001 runs the computer programs stored in the memory 1002 to realize various functions, as follows: Obtain the sampled traffic path within the area to be reconstructed. The sampled traffic path consists of at least one road segment and includes the starting traffic area and the ending traffic area. Based on the origin traffic area and the destination traffic area, the sampled traffic paths are clustered to obtain the target traffic paths between the origin traffic area and the destination traffic area. Obtain the traffic flow at checkpoints obtained from traffic flow detection of the target road segment within the area to be reconstructed; Based on the target traffic path, the traffic flow at the checkpoint is allocated using maximum entropy to obtain the path flow of the target traffic path with respect to the target road segment; Based on the target traffic path and path flow, path reconstruction is performed on the area to be reconstructed to obtain the path distribution and path flow within the area to be reconstructed. Specific implementation details for each of these operations can be found in the preceding embodiments and will not be repeated here.

[0119] As can be seen from the above, the computer device in this embodiment of the application can obtain sampled traffic paths in the area to be reconstructed. The sampled traffic paths consist of at least one road segment and include a starting traffic area and an ending traffic area. Based on the starting traffic area and the ending traffic area, the sampled traffic paths are subjected to path clustering processing to obtain the target traffic path between the starting traffic area and the ending traffic area. The checkpoint traffic flow obtained by traffic flow detection of the target road segment in the area to be reconstructed is obtained. Based on the target traffic path, the checkpoint traffic flow is allocated with maximum entropy to obtain the path flow of the target traffic path with respect to the target road segment. Based on the target traffic path and the path flow, the path is reconstructed in the area to be reconstructed to obtain the path distribution and path flow in the area to be reconstructed.

[0120] This application embodiment performs path clustering on sampled traffic paths to obtain popular target traffic paths in the area to be reconstructed. Then, it reconstructs the target traffic paths based on the traffic flow at the checkpoints set up along the target traffic paths. Based on the sampled target traffic paths and the total traffic flow obtained through checkpoint detection, it expands the sampling to obtain the path distribution and path flow in the area to be reconstructed, making the path reconstruction results consistent with the actual traffic flow in the area to be reconstructed, and the path reconstruction results are more accurate.

[0121] According to one aspect of this application, a computer program product is provided, comprising a computer program containing computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the methods provided in the various optional implementations of the above embodiments.

[0122] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by a computer program, or by a computer program controlling related hardware. The computer program can be stored in a computer-readable storage medium and loaded and executed by a processor.

[0123] Therefore, embodiments of this application provide a computer-readable storage medium storing a computer program that can be loaded by a processor to execute any of the path reconstruction methods provided in embodiments of this application.

[0124] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0125] The computer-readable storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0126] Since the computer program stored in the computer-readable storage medium can execute any of the path reconstruction methods provided in the embodiments of this application, it can achieve the beneficial effects that any of the path reconstruction methods provided in the embodiments of this application can achieve, as detailed in the preceding embodiments, and will not be repeated here.

[0127] The path reconstruction method, apparatus, computer device, and computer-readable storage medium provided in the embodiments of this application have been described in detail above. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A path reconstruction method, characterized in that, include: Obtain sampled traffic paths within the area to be reconstructed. Each sampled traffic path consists of at least one road segment and includes a starting traffic area and an ending traffic area. Based on the starting traffic area and the ending traffic area, the sampled traffic path is subjected to path clustering processing to obtain the target traffic path between the starting traffic area and the ending traffic area; Obtain the traffic flow at the checkpoints by performing traffic flow detection on the target road segment within the area to be reconstructed; Based on the target traffic path, the maximum entropy flow allocation is performed on the checkpoint traffic flow to obtain the path flow of the target traffic path with respect to the target road segment. This includes: each target traffic path contains at least one target road segment; based on the path flow of the target traffic path with respect to the at least one target road segment, an adjustment parameter for the path flow is calculated; the path flow is adjusted based on the adjustment parameter to obtain an initial adjusted path flow; based on the initial adjusted path flow and the target traffic path, the maximum entropy flow allocation is performed on the checkpoint traffic flow to obtain the adjusted path flow of the target traffic path with respect to the target road segment; the adjusted path flow is used as the path flow, and the process of calculating the adjustment parameter for the path flow based on the path flow of the target traffic path with respect to the at least one target road segment is repeated until the adjustment parameter meets a preset condition. Based on the target traffic path and the path flow, the path to be reconstructed area is reconstructed to obtain the path distribution and path flow within the area to be reconstructed.

2. The method according to claim 1, characterized in that, The step of calculating adjustment parameters for the path flow based on the path flow of the target traffic path with respect to at least one target road segment includes: From multiple target road segments, a first target road segment and a second target road segment are determined, wherein the first target road segment and the second target road segment are adjacent road segments; The adjustment parameter is calculated based on the ratio between the path flow of the target traffic path with respect to the first target road segment and the path flow with respect to the second target road segment.

3. The method according to claim 1, characterized in that, The step of performing path clustering processing on the sampled traffic paths based on the origin traffic area and the destination traffic area to obtain the target traffic path between the origin traffic area and the destination traffic area includes: Calculate the path similarity between each sampled traffic path and other sampled traffic paths among multiple sampled traffic paths between each pair of origin traffic areas and destination traffic areas; Based on the path similarity, path clustering is performed on multiple sampled traffic paths between each pair of starting traffic areas and ending traffic areas to obtain clustering results; The target traffic route is determined based on the clustering results.

4. The method according to claim 3, characterized in that, The path clustering process is performed on multiple sampled traffic paths between each pair of origin and destination traffic areas based on the path similarity, resulting in clustering results, including: Based on the path similarity between each sampled traffic path between each pair of origin traffic areas and destination traffic areas and other sampled traffic paths between each pair of origin traffic areas and destination traffic areas, the path feature information of each sampled traffic path is obtained. Based on the path feature information, path clustering is performed on the multiple sampled traffic paths to obtain the clustering results.

5. The method according to claim 4, characterized in that, The clustering results include multiple clusters, and determining the target traffic path based on the clustering results includes: The number of sampled traffic paths contained in each cluster is determined based on the clustering results; The filtering threshold is determined based on the number of paths; The target traffic path is obtained by filtering from the sampled traffic paths according to the filtering threshold.

6. The method according to any one of claims 1-5, characterized in that, The acquisition of multiple sampled traffic paths within the area to be reconstructed includes: Obtain candidate sampled traffic paths within the area to be reconstructed, wherein the candidate sampled traffic paths include candidate origin traffic areas and candidate destination traffic areas; Based on the candidate sampled traffic paths, calculate the first number of times the candidate starting point traffic area is set as the starting point and the second number of times the candidate ending point traffic area is set as the ending point. A starting traffic area is selected from the candidate starting traffic areas based on the first number of times, and a destination traffic area is selected from the candidate destination traffic areas based on the second number of times; Based on the origin traffic area and the destination traffic area, the sampled traffic path is selected from the candidate sampled traffic paths.

7. A path reconstruction device, characterized in that, include: The path acquisition unit is used to acquire sampled traffic paths within the area to be reconstructed. The sampled traffic paths consist of at least one road segment and include a starting traffic area and an ending traffic area. A clustering unit is used to perform path clustering processing on the sampled traffic path based on the starting traffic area and the ending traffic area to obtain the target traffic path between the starting traffic area and the ending traffic area; The traffic flow acquisition unit is used to acquire the traffic flow at the checkpoint obtained by traffic flow detection of the target road segment in the area to be reconstructed. The allocation unit is used to perform maximum entropy flow allocation on the traffic flow at the checkpoint based on the target traffic path, so as to obtain the path flow of the target traffic path with respect to the target road segment; The reconstruction unit is used to reconstruct the path in the area to be reconstructed based on the target traffic path and the path flow, so as to obtain the path distribution and path flow in the area to be reconstructed. The allocation unit includes: The calculation subunit is used to calculate the adjustment parameters for the path flow based on the path flow of the target traffic path with respect to at least one target road segment it contains; An adjustment subunit is used to adjust the path flow based on the adjustment parameters to obtain the initial adjusted path flow. The traffic flow allocation subunit is used to perform maximum entropy traffic flow allocation on the checkpoint traffic flow based on the initial adjusted path flow and the target traffic path, so as to obtain the adjusted path flow of the target traffic path with respect to the target road segment. The loop subunit is used to take the adjusted path flow as the path flow and return to execute the step of calculating the adjustment parameters of the path flow based on the path flow of the target traffic path with respect to at least one target road segment, until the adjustment parameters meet the preset conditions.

8. The apparatus according to claim 7, characterized in that, The computing subunit includes: The selection module is used to determine a first target road segment and a second target road segment from multiple target road segments, wherein the first target road segment and the second target road segment are adjacent road segments; The parameter calculation module is used to calculate the adjustment parameter based on the ratio between the path flow of the target traffic path with respect to the first target road segment and the path flow with respect to the second target road segment.

9. The apparatus according to claim 7, characterized in that, The clustering unit includes: The similarity calculation subunit is used to calculate the path similarity between each sampled traffic path and other sampled traffic paths in each pair of origin traffic areas and destination traffic areas. The path clustering subunit is used to perform path clustering processing on multiple sampled traffic paths between each pair of starting traffic areas and ending traffic areas based on the path similarity, and obtain the clustering result. The target path determination subunit is used to determine the target traffic path based on the clustering results.

10. The apparatus according to claim 9, characterized in that, The path clustering subunit includes: The feature information acquisition module is used to obtain the path feature information of each sampled traffic path based on the path similarity between each sampled traffic path between each pair of origin traffic areas and destination traffic areas and other sampled traffic paths between each pair of origin traffic areas and destination traffic areas. The clustering processing module is used to perform path clustering processing on the multiple sampled traffic paths based on the path feature information to obtain clustering results.

11. The apparatus according to claim 10, characterized in that, The clustering result includes multiple clusters, and the target path determination sub-unit includes: The quantity determination module is used to determine the number of sampled traffic paths contained in each cluster based on the clustering results; The threshold determination module is used to determine the filtering threshold based on the number of paths; The filtering module is used to filter the target traffic path from the sampled traffic paths according to the filtering threshold.

12. The apparatus according to any one of claims 7 to 11, characterized in that, The path acquisition unit includes: The candidate path acquisition subunit is used to acquire candidate sampled traffic paths within the area to be reconstructed, wherein the candidate sampled traffic paths include candidate origin traffic areas and candidate destination traffic areas; The calculation subunit is used to calculate, based on the candidate sampled traffic path, the first number of times the candidate starting point traffic area is set as the starting point and the second number of times the candidate ending point traffic area is set as the ending point. A selection subunit is configured to select a starting traffic area from the candidate starting traffic areas based on the first number of times, and to select a destination traffic area from the candidate destination traffic areas based on the second number of times. The route filtering subunit is used to filter the sampled traffic routes from the candidate sampled traffic routes based on the origin traffic area and the destination traffic area.

13. A computer device, characterized in that, It includes a memory and a processor; the memory stores a computer program, and the processor is used to run the computer program in the memory to perform the path reconstruction method according to any one of claims 1 to 6.

14. A computer-readable storage medium, characterized in that, The computer-readable storage medium is used to store a computer program, which is loaded by a processor to perform the path reconstruction method according to any one of claims 1 to 6.

15. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the path reconstruction method according to any one of claims 1 to 6.

Images