A method and system for displaying traffic road network situation data

Abstract

The present application relates to the field of data processing technology, and provides a method and system for displaying traffic network situation data. By collaboratively processing traffic network situation data in multiple steps, the traffic situation perception and presentation capabilities are comprehensively improved. From the acquisition of multi-source heterogeneous data to the final dynamic adjustment of the grid granularity according to the deduction error, a closed-loop optimization system is formed, which can provide accurate, real-time and personalized traffic network situation information to traffic management departments and travelers, assist in decision-making and improve travel efficiency, and has significant innovation and practicality in the field of intelligent transportation.

Description

Technical Field

[0001] This application relates to the technical field of data processing, and particularly to a method and system for displaying traffic road network situation data. Background Art

[0002] In the existing traffic road network situation data display technology, a common practice is to analyze traffic conditions using a single or a few data sources. For example, only relying on traffic flow data to judge traffic congestion and perform simple visual displays, without considering the complex relationships between different road segments and regions and real-time situation feedback and adjustment. Most of them do not solve how to achieve efficient optimization of traffic data processing and dynamic display of traffic situations. Summary of the Invention

[0003] In view of the deficiencies of the prior art, this application provides a method and system for displaying traffic road network situation data.

[0004] In a first aspect, this application provides a method for displaying traffic road network situation data, the method including: obtaining multi-source heterogeneous data including traffic flow data, radar trajectory data, and situation propagation data in a traffic road network, performing real-time interpolation on the missing multi-source heterogeneous data through an adversarial generative network, and performing spatio-temporal synchronization on the interpolated multi-source heterogeneous data to generate a traffic road network topology;

[0005] Decomposing the traffic road network topology into grid cells based on grid granularity, calculating the spatio-temporal conduction entropy between adjacent grid cells, and determining the basic field strength of each grid cell according to the spatio-temporal conduction entropy between adjacent grid cells to generate a situation field strength map of the traffic road network;

[0006] Extracting the situation feature tensor of the situation field strength map, the situation feature tensor including situation layer features, event layer features, and deduction layer features, and performing visual encoding on the situation feature tensor according to the basic field strength of each grid cell to determine a hierarchical rendering instruction;

[0007] Obtaining eye movement tracking data to dynamically adjust the basic field strength of each grid cell, and performing semantic description and position annotation on the hierarchical rendering instruction through natural language processing and annotation positioning algorithms, performing multi-layer situation deduction on the traffic road network according to the hierarchical rendering instruction, and dynamically adjusting the grid granularity according to the error of the multi-layer situation deduction.

[0008] As an optional implementation manner, the generation logic of the traffic road network topology includes:

[0009] Obtaining multi-source heterogeneous data including traffic flow data, radar trajectory data, and situation propagation data from the traffic road network;

[0010] Performing real-time interpolation on the missing multi-source heterogeneous data through an adversarial generative network;

[0011] Synchronize the interpolated multi-source heterogeneous data according to the constructed spatial reference points and temporal reference points;

[0012] For the multi-source heterogeneous data after spatio-temporal synchronization, determine the connection relationships of nodes and the weights of edges in the traffic road network through a topology algorithm, and generate a traffic road network topology.

[0013] As an optional implementation manner, the generation logic of the situation field strength map includes:

[0014] Decompose the traffic road network topology into grid cells based on the grid granularity, and calculate the spatio-temporal conduction entropy between adjacent grid cells;

[0015] Use the spatio-temporal conduction entropy between adjacent grid cells as weights, and determine the basic field strength of each grid cell by means of weighted average;

[0016] Integrate the basic field strength and spatio-temporal conduction entropy of each grid cell to generate a situation field strength map, and visually display the situation field strength map.

[0017] As an optional implementation manner, the decomposition sub-logic of the grid cells includes:

[0018] Extract features of the traffic road network topology through a convolutional neural network to identify the traffic road network type;

[0019] Determine the grid granularity according to the traffic road network type;

[0020] Decompose the traffic road network topology into grid cells based on the grid granularity, and determine the boundaries of each grid cell according to its position in the traffic road network.

[0021] As an optional implementation manner, the calculation sub-logic of the spatio-temporal conduction entropy between adjacent grid cells includes:

[0022] Extract the multi-source heterogeneous data of each grid cell from the traffic road network topology;

[0023] Analyze the causal relationship and spatial correlation of the multi-source heterogeneous data between adjacent grid cells to obtain a spatio-temporal correlation relationship;

[0024] Calculate the spatio-temporal conduction entropy between adjacent grid cells based on the spatio-temporal correlation relationship.

[0025] As an optional implementation manner, the extraction logic of the situation feature tensor includes:

[0026] Divide the situation field strength map into a situation layer, an event layer, and a deduction layer;

[0027] Configure multi-layer extraction units to extract the situation layer features, event layer features, and deduction layer features of the situation field strength map in parallel through the multi-layer extraction units;

[0028] The situation layer features, event layer features, and deduction layer features are fused through an attention mechanism to obtain a situation feature tensor.

[0029] As an alternative implementation, the determination logic of the hierarchical rendering instruction includes:

[0030] Analyze the mapping relationship between the basic field strength of each grid cell and the situation feature tensor through a neural network;

[0031] Visually encode the situation layer features, event layer features, and deduction layer features respectively, and adjust the parameters of the visual encoding in combination with the basic field strength of each grid cell;

[0032] Generate rendering instructions for the situation layer, event layer, and deduction layer based on the results of the visual encoding and the mapping relationship between the basic field strength of each grid cell and the situation feature tensor to determine the hierarchical rendering instruction.

[0033] As an alternative implementation, the adjustment logic of the basic field strength of each grid cell includes:

[0034] Real-time obtain the user's eye movement tracking data, which includes the fixation point position, fixation time, and saccade path, and map the eye movement tracking data to the grid cells;

[0035] Identify the grid cells concerned by the user through a clustering algorithm to obtain the attention degree of each grid cell;

[0036] Determine the adjustment direction and adjustment amplitude of the basic field strength of each grid cell according to the attention degree of each grid cell;

[0037] Obtain the adjusted basic field strength of each grid cell according to the adjustment direction and adjustment amplitude of the basic field strength of each grid cell and the current basic field strength of each grid cell.

[0038] As an alternative implementation, the adjustment logic of the grid granularity includes:

[0039] During the multi-layer situation deduction of the traffic road network according to the hierarchical rendering instruction, the error magnitude and error change trend between the deduction results of the situation layer, event layer, and deduction layer and the actual traffic conditions are recorded in real time through the mean square error;

[0040] Through a fuzzy logic reasoning method, determine whether to adjust the grid granularity and the adjustment direction of the grid granularity according to the error magnitude, error change trend, and actual traffic conditions;

[0041] Adjust the grid granularity according to the adjustment direction of the grid granularity, and perform multi-layer situation deduction again. Compare the error magnitudes before and after adjusting the grid granularity to determine whether to readjust the grid granularity.

[0042] In a second aspect, the present application provides a traffic road network situation data display system, which includes: a topology generation module, a map generation module, a hierarchical rendering module, and a feedback adjustment module;

[0043] The topology generation module is used to obtain multi-source heterogeneous data including traffic flow data, radar trajectory data, and incident propagation data in the traffic road network, perform real-time interpolation on the missing multi-source heterogeneous data through an adversarial generation network, and perform spatio-temporal synchronization on the interpolated multi-source heterogeneous data to generate a traffic road network topology;

[0044] The map generation module is used to decompose the traffic road network topology into grid cells based on grid granularity, calculate the spatio-temporal conduction entropy between adjacent grid cells, and determine the basic field strength of each grid cell according to the spatio-temporal conduction entropy between adjacent grid cells to generate a situation field strength map of the traffic road network;

[0045] The hierarchical rendering module is used to extract the situation feature tensor of the situation field strength map. The situation feature tensor includes situation layer features, event layer features, and deduction layer features, and perform visual encoding on the situation feature tensor according to the basic field strength of each grid cell to determine the hierarchical rendering instruction;

[0046] The feedback adjustment module is used to obtain eye movement tracking data to dynamically adjust the basic field strength of each grid cell, semantically describe and position label the hierarchical rendering instruction through natural language processing and annotation positioning algorithms, perform multi-layer situation deduction on the traffic road network according to the hierarchical rendering instruction, and dynamically adjust the grid granularity according to the error of the multi-layer situation deduction.

[0047] Compared with the prior art, the beneficial effects of the present application are: by collaboratively processing traffic road network situation data in multiple steps, the traffic situation perception and presentation capabilities are comprehensively improved. From obtaining multi-source heterogeneous data to finally dynamically adjusting the grid granularity according to the deduction error, a closed-loop optimization system is formed, which can provide accurate, real-time, and personalized traffic road network situation information for traffic management departments and travelers, etc., assist in decision-making and improve travel efficiency, and has significant innovation and practicality in the field of intelligent transportation.

[0048] Integrating traffic flow data, radar trajectory data, and incident propagation data, comprehensively reflecting the actual situation of the traffic road network. The adversarial generation network interpolates missing data in real time to ensure data integrity. Spatio-temporal synchronization unifies the spatio-temporal reference of the data, providing a reliable basis for subsequent analysis. The generated traffic road network topology accurately presents the node connection relationship and edge weights, laying a foundation for in-depth analysis of traffic situations.

[0049] Based on the grid granularity to decompose the traffic road network topology, calculate the spatio-temporal conduction entropy to determine the basic field strength. The generated situation field strength map can visually display the situation strength and mutual relationship of each grid unit. By means of differential grid division to adapt to different types of traffic road networks, it helps to analyze the details of traffic situations more accurately and provides key data support for subsequent feature extraction and visualization.

[0050] Extract multi-level features of the situation field strength map to form a situation feature tensor, and determine the hierarchical rendering instructions based on the visual coding of the basic field strength. This enables the traffic situation data to be presented in a hierarchical visualization form, with different levels of information clearly distinguished. Users can focus on specific levels according to their needs, improving the efficiency of information acquisition. Moreover, the basic field strength affects the visual coding, highlighting the situation in key areas.

[0051] Obtain eye movement tracking data to adjust the basic field strength, realize personalized situation display, and highlight the areas of user concern. Process the hierarchical rendering instructions through natural language processing and annotation positioning algorithms for easy understanding and application. Dynamically adjust the grid granularity according to the multi-level situation deduction error to optimize the accuracy of situation deduction, enabling the system to adapt to the changing traffic conditions and continuously provide high-quality situation information. Description of the Drawings

[0052] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. Among them:

[0053] Figure 1 It is a flowchart of a method for displaying traffic road network situation data provided by an embodiment of the present application;

[0054] Figure 2 It is a logic diagram of grid unit decomposition for a method for displaying traffic road network situation data provided by an embodiment of the present application;

[0055] Figure 3 It is a system module diagram of a system for displaying traffic road network situation data provided by an embodiment of the present application. Detailed Embodiments

[0056] In order to make the objectives, technical solutions, and advantages of the embodiments of the present application more obvious and understandable, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the specification. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments.

[0057] Embodiment 1

[0058] As Figure 1As shown in the figure, the embodiment of the present application provides a method flowchart of a traffic road network situation data display method, and the method includes:

[0059] S1. Obtain multi-source heterogeneous data including traffic flow data, radar trajectory data, and incident propagation data in the traffic road network, perform real-time imputation on the missing multi-source heterogeneous data through an adversarial generative network, and perform spatio-temporal synchronization on the imputed multi-source heterogeneous data to generate a traffic road network topology.

[0060] Specifically, the generation logic of the traffic road network topology includes:

[0061] Obtain multi-source heterogeneous data including traffic flow data, radar trajectory data, and incident propagation data from the traffic road network;

[0062] Perform real-time imputation on the missing multi-source heterogeneous data through an adversarial generative network;

[0063] Perform spatio-temporal synchronization on the imputed multi-source heterogeneous data according to the constructed spatial reference point and temporal reference point;

[0064] Determine the connection relationship between nodes and the weights of edges in the traffic road network through a topology algorithm for the spatio-temporally synchronized multi-source heterogeneous data to generate a traffic road network topology.

[0065] Traffic flow data can reflect the traffic conditions of vehicles on the road, such as traffic volume and speed, etc. Radar trajectory data can accurately track the real-time positions and movement trajectories of vehicles, while incident propagation data contains information related to special events such as traffic accidents and road construction. Only by integrating these multi-source heterogeneous data can the actual situation of the traffic road network be comprehensively and accurately described, providing a rich data basis for subsequent analysis and processing; install induction coils and cameras at key positions on the road (such as intersections and in the middle of sections, etc.). Among them, the induction coil detects the magnetic field change generated when a vehicle passes through through the principle of electromagnetic induction, so as to obtain data such as traffic volume and vehicle speed, while the camera uses image recognition technology to analyze the captured video stream, identify vehicles and count relevant information.

[0066] Through a millimeter-wave radar, which emits millimeter waves and receives the signals reflected by vehicles, the speed and position information of the vehicles are calculated by analyzing the frequency change of the signals (Doppler effect), thereby obtaining radar trajectory data. The millimeter-wave radar is usually installed at a high place to obtain a wider monitoring range; while the acquisition of incident propagation data is achieved by obtaining information on traffic accidents and road construction and other incidents released by the official, or by using natural language processing technology to collect traffic-related text information from platforms such as traffic information websites, and extracting the incident information therein. For example, texts containing accident information are identified through text classification algorithms, and key information such as the location, time, and type of the accident are further extracted; multi-source data fusion can provide a more comprehensive traffic network situation awareness. Traffic flow data reflects the traffic operation status from a macroscopic level, radar trajectory data provides the precise positions of vehicles at a microscopic level, and incident propagation data can reveal special events affecting traffic. The three complement each other, making the data more complete and accurate.

[0067] In the actual data acquisition process, due to reasons such as equipment failures and communication interruptions, data missing inevitably occurs. Missing data will affect the accuracy and integrity of subsequent data analysis, leading to misjudgments of the traffic network situation. Through a generative adversarial network, real-time imputation of missing multi-source heterogeneous data is performed. The generator of the generative adversarial network adopts a fully connected neural network structure. Input a random noise vector and the known part of the multi-source heterogeneous data. Through the non-linear transformation of multiple layers of neurons, the estimated value of the missing data is output. The discriminator is also a fully connected neural network. Its input is the complete multi-source heterogeneous data, including real data and the imputed data generated by the generator, and a probability value is output, indicating the possibility that the input data is real data. A large amount of historical multi-source heterogeneous data is used to train the generative adversarial network. During the training process, the generator attempts to generate realistic missing data to deceive the discriminator, while the discriminator tries to distinguish between real data and generated data. Through continuous iterative training, finally the generator can generate imputation values of missing data close to the real data distribution. For example, during the training process, the random gradient descent algorithm is used to update the parameters of the generator and the discriminator, and the cross-entropy loss function is used as the loss function to measure the difference between the generated data and the real data. During the real-time data acquisition process, once data missing is detected, the known part of the data is immediately input into the trained generator, and the generator quickly outputs the imputation value of the missing data to achieve real-time imputation; thereby improving the integrity and accuracy of the data, reducing the analysis error caused by data missing. The imputed data generated by the generative adversarial network can better retain the characteristics and distribution laws of the original data, making the subsequent analysis results based on these data more reliable.

[0068] Multi-source heterogeneous data may exhibit inconsistencies in space and time. Spatiotemporal synchronization can unify the spatiotemporal benchmarks of multi-source heterogeneous data, enabling them to be analyzed and processed within the same spatiotemporal framework, thereby accurately reflecting the true situation of the traffic road network. Based on the map in the geographic information system, key landmarks in the traffic road network (such as large buildings and important intersections) are selected as spatial reference points. By measuring the coordinates of these key landmarks in different data acquisition methods and using coordinate transformation algorithms, such as the seven-parameter transformation model, the spatial coordinates of different data are uniformly transformed into a coordinate system referenced by these spatial reference points. An atomic clock is used as the time synchronization source to provide a unified time benchmark, ensuring the accuracy and consistency of the data acquisition time. Meanwhile, during the data acquisition process, the acquisition timestamp of each data point is recorded.

[0069] For the interpolated multi-source heterogeneous data, according to their timestamps and spatial coordinates, they are uniformly mapped into a spatiotemporal coordinate system based on the constructed spatial and time reference points. For example, for traffic flow data and radar trajectory data, according to their acquisition time and spatial position, they are matched and aligned in the unified spatiotemporal coordinate system to ensure that the data at the same moment and the same position can accurately correspond. Thus, the consistency of multi-source heterogeneous data in the spatiotemporal dimension is achieved, eliminating data chaos and errors caused by spatiotemporal inconsistencies. The unified spatiotemporal benchmark makes the subsequent analysis and understanding of the traffic road network situation more accurate and intuitive, and can better grasp the changing rules of traffic flow in time and space.

[0070] The topology of the traffic road network is the key to understanding the operation of traffic flow and situation analysis. By determining the connection relationships of nodes and the weights of edges, the complex traffic road network can be abstracted into an analyzable graph structure, facilitating subsequent in-depth research and visual display of traffic situations. According to the actual situation of the traffic road network, intersections and road section endpoints are used as nodes. By analyzing the multi-source heterogeneous data after spatiotemporal synchronization and through geographic information data, the position information of these nodes is identified. For example, the coordinates of intersections are extracted from GIS map data, and the start and end positions of road sections are determined from traffic flow data, thereby determining the specific positions of the nodes. Based on the actual layout of the traffic road network and geographic information, the connection relationships between nodes are determined. For example, adjacent intersections are connected by road sections, and the two ends of a road section correspond to two nodes respectively. Using the shortest path algorithm and combining traffic rules, such as one-way street restrictions, the connection relationships between nodes are constructed.

[0071] The weights of the edges are calculated according to different traffic indicators. For example, when taking traffic flow as the weight, the number of vehicles passing through the road section (edge) within a unit time is counted. When taking the passing speed as the weight, the average vehicle speed of the road section is calculated through radar trajectory data and traffic flow data. Multiple indicators can also be comprehensively considered. For example, traffic flow and passing speed are weighted and combined to obtain a comprehensive weight value. According to the determined connection relationships of the nodes and the weights of the edges, a traffic road network topology is generated through Dijkstra's algorithm, thereby transforming the complex traffic road network into a graph structure that is easy to understand and analyze, providing an intuitive and effective tool for traffic situation analysis. By reasonably setting the weights of the edges, the importance of different road sections in the traffic road network and the traffic operation conditions can be reflected, which helps traffic managers quickly understand traffic bottlenecks and congested road sections. The generated traffic road network topology provides a basic framework for subsequent steps such as decomposing it into grid cells and calculating the spatio-temporal conduction entropy between adjacent grid cells. An accurate topology structure can make subsequent grid division and analysis more in line with the actual traffic situation, improving the accuracy and effectiveness of the entire traffic road network situation data display method.

[0072] S2. Decompose the traffic road network topology into grid cells based on the grid granularity, calculate the spatio-temporal conduction entropy between adjacent grid cells, and determine the basic field strength of each grid cell according to the spatio-temporal conduction entropy between adjacent grid cells to generate a situation field strength map of the traffic road network.

[0073] Specifically, the generation logic of the situation field strength map includes:

[0074] Decompose the traffic road network topology into grid cells based on the grid granularity and calculate the spatio-temporal conduction entropy between adjacent grid cells;

[0075] Take the spatio-temporal conduction entropy between adjacent grid cells as the weight and determine the basic field strength of each grid cell by means of weighted average;

[0076] Integrate the basic field strength and spatio-temporal conduction entropy of each grid cell to generate a situation field strength map, and visually display the situation field strength map.

[0077] By decomposing the traffic road network topology into grid cells and calculating the spatio-temporal conduction entropy between adjacent cells, it is possible to obtain the relative situation influence information of each grid cell in the traffic road network, providing a data basis for subsequent determination of the basic field strength and generation of the situation field strength map. According to the determined grid granularity, the traffic road network topology is divided into grids in the GIS environment to determine the boundaries of each grid cell. Then, multi-source heterogeneous data of each grid cell are extracted from the traffic road network topology, the causal relationship and spatial correlation between adjacent grid cells are analyzed, and the spatio-temporal conduction entropy between adjacent grid cells is calculated based on the spatio-temporal correlation relationship. In actual operation, parallel computing technology is used to calculate the spatio-temporal conduction entropy of multiple adjacent grid cell pairs simultaneously to improve the calculation efficiency. Thus, the spatio-temporal conduction relationship between each grid cell and its adjacent cells in the traffic road network is comprehensively obtained, providing rich and accurate data support for subsequent generation of the situation field strength map and helping to more accurately reflect the situation characteristics of the traffic road network.

[0078] The basic field strength of each grid cell reflects the relative importance and traffic situation intensity of the grid cell in the traffic road network. By taking the spatio-temporal conduction entropy between adjacent grid cells as the weight for weighted average, the traffic situation transmission relationship between the grid cell and its adjacent cells can be comprehensively considered, so as to more reasonably determine its basic field strength. For each grid cell, collect the spatio-temporal conduction entropy between it and all adjacent grid cells. Suppose the grid cell has adjacent grid cells, and the spatio-temporal conduction entropy between the adjacent grid cell and the grid cell is . Then, the basic field strength of the grid cell is calculated by the following formula:

[0079] ;

[0080] In the formula, represents the basic field strength of the grid cell , represents the number of adjacent grid cells, is the spatio-temporal conduction entropy between the adjacent grid cell and the grid cell , represents a certain traffic-situation-related eigenvalue of the adjacent grid cell , such as traffic flow and average vehicle speed.

[0081] In actual calculations, first determine the eigenvalues related to traffic situations for the calculations, and then calculate the basic field strength of each grid cell according to the above formula; thereby making the determination of the basic field strength more scientific and reasonable, fully considering the spatio-temporal correlation relationship between adjacent grid cells, being able to more accurately reflect the actual situation strength of each grid cell in the traffic road network, and providing key support for generating an accurate situation field strength map.

[0082] Integrate the basic field strength and spatio-temporal conduction entropy to generate a situation field strength map, which can present the situation information of each grid cell in the traffic road network in an intuitive and visual way; based on a geographic information system, take each grid cell as an element in the situation field strength map, and in the situation field strength map, layout according to the position of the grid cell. The color, size or other visual attributes of each grid cell can be encoded according to the values of its basic field strength and spatio-temporal conduction entropy. For example, grid cells with a large basic field strength (obtained by comparing with the field strength threshold) are represented by a darker color, and grid cells with a large spatio-temporal conduction entropy (obtained by comparing with the entropy value threshold) are represented by a larger size. In this way, the basic field strength and spatio-temporal conduction entropy information of each grid cell are intuitively displayed in the map.

[0083] Display the generated situation field strength map through visualization software, providing various display methods, such as static map display and dynamic animation display (displaying the evolution of traffic situations over time), etc., and at the same time adding interactive functions. For example, when the mouse hovers over a certain grid cell, display the detailed basic field strength, spatio-temporal conduction entropy and related traffic data information of that cell, facilitating users to deeply understand the situation of each grid cell; thereby presenting the situation information of the traffic road network in an intuitive and visual way, greatly improving the readability and understandability of traffic situation data, and helping traffic managers make decisions quickly and take corresponding traffic management measures.

[0084] Furthermore, as Figure 2 shown, the decomposition sub-logic of grid cells includes:

[0085] Extract features of the traffic road network topology through a convolutional neural network to identify the traffic road network type;

[0086] Determine the grid granularity according to the traffic road network type;

[0087] Decompose the traffic road network topology into grid cells based on the grid granularity, and determine the boundary of each grid cell according to its position in the traffic road network.

[0088] Different types of traffic road networks, such as highways, urban arterial roads, and branch roads, have different traffic characteristics and structural features. Accurately identifying the type of traffic road network helps to adopt appropriate grid division strategies according to its characteristics in the subsequent process, improving the accuracy of traffic situation analysis. By using a convolutional neural network, the traffic road network topology is transformed into an image form. For example, taking road network nodes as pixel points, different colors or gray values are assigned to the pixels according to the connection relationship between nodes and the weights of edges. Then, this image data is input into the convolutional neural network. The convolutional neural network contains multiple convolutional layers, pooling layers, and fully connected layers. The convolutional layers extract local features in the image through convolutional kernels of different sizes. The pooling layers perform downsampling on the output of the convolutional layers to reduce the data volume and retain important features. The fully connected layers integrate the features after multiple convolutions and poolings and output the probability distribution of the traffic road network type through the softmax function, thereby determining the traffic road network type. During the training process, a large number of traffic road network topologies of known types are used for training, and the model parameters are continuously adjusted through the backpropagation algorithm to improve the classification accuracy. Thus, it is possible to quickly and accurately identify the type of traffic road network, providing a basis for determining the grid granularity in the subsequent process, making the grid division more in line with the actual situation of different types of traffic road networks, and thus more effectively capturing traffic situation information.

[0089] There are differences in aspects such as traffic flow, speed changes, and spatial scales among different types of traffic road networks. For example, highways have large traffic flow and high speed, and require a larger grid granularity to summarize the overall traffic situation. While branch roads have complex traffic conditions and relatively small traffic flow, and require a smaller grid granularity to describe the local traffic conditions in detail. Therefore, determining the grid granularity according to the type of traffic road network can more effectively utilize computing resources and accurately reflect the traffic characteristics of different road networks. For highways, based on the number of lanes, designed speed, and historical traffic flow data, a larger grid granularity is set, such as a square grid with a side length of every 5 kilometers or 10 kilometers. For urban arterial roads, considering the changes in traffic flow and road length, a moderate grid granularity is set, such as a grid with a side length of every 1 kilometer or 2 kilometers. For branch roads, due to their narrow roads and complex traffic conditions, a smaller grid granularity is adopted, such as a grid with a side length of every 200 meters or 500 meters. Thus, the grid division becomes more scientific and reasonable, which can reduce unnecessary computational workload and improve data processing efficiency while ensuring an accurate description of the traffic situation. At the same time, setting the grid granularity according to the characteristics of different types of road networks can more accurately reflect traffic changes.

[0090] Decomposing the traffic road network topology into grid cells can divide the complex traffic road network into relatively independent and interrelated small areas, facilitating a detailed analysis of the traffic situation in each area. Determining the boundaries of each grid cell is to clarify the scope of the traffic road network covered by each grid cell, so as to accurately extract the traffic data in this area. Based on the determined grid granularity, grid division of the traffic road network topology is carried out in the geographical information system environment. For example, if the grid granularity is a square grid with a side length of 1 kilometer, then on the GIS map, grid lines are drawn at intervals of 1 kilometer in the horizontal and vertical directions, and the traffic road network topology is covered under the grid. For each grid cell, through the spatial analysis function of GIS, the intersection situation between its boundary and the traffic road network is determined, so as to clarify the road network segments and nodes contained in this grid cell, and thus determine the boundary of the grid cell. The effective segmentation of the traffic road network is realized, enabling the analysis of the traffic situation to be refined to each grid cell, improving the accuracy of the analysis. At the same time, the clear grid cell boundaries contribute to accurately extracting the traffic data of each cell, providing an accurate data source for subsequent calculation of the spatio-temporal conduction entropy and determination of the basic field strength.

[0091] Furthermore, the calculation sub-logic of the spatio-temporal conduction entropy between adjacent grid cells includes:

[0092] Extracting the multi-source heterogeneous data of each grid cell from the traffic road network topology;

[0093] Analyzing the causal relationship and spatial correlation of the multi-source heterogeneous data between adjacent grid cells to obtain the spatio-temporal correlation relationship;

[0094] Calculating the spatio-temporal conduction entropy between adjacent grid cells based on the spatio-temporal correlation relationship.

[0095] The spatio-temporal conduction entropy is used to measure the degree of mutual influence of the traffic situation between adjacent grid cells in the spatio-temporal dimension; according to the previously determined boundaries of the grid cells, the multi-source heterogeneous data within each grid cell is extracted from the traffic road network topology; thus, the multi-source data of each grid cell is comprehensively obtained, providing rich data support for subsequent analysis of the spatio-temporal correlation relationship between adjacent grid cells, and being able to more accurately reflect the interaction of the traffic situation between grid cells.

[0096] Understanding the causal relationship and spatial correlation of multi-source heterogeneous data between adjacent grid cells can deeply understand the propagation and influence mechanism of traffic situations between different grid cells. The causal relationship can reveal whether the change in the traffic state of one grid cell will trigger corresponding changes in adjacent grid cells, while spatial correlation reflects the similarity degree of traffic states between adjacent grid cells in space. The spatio-temporal correlation relationship obtained by integrating these two aspects is the key factor for calculating spatio-temporal conduction entropy. For the time series data of adjacent grid cells through the Granger causality test method, such as the change sequence of traffic flow over time, by constructing a vector autoregressive model, it is tested whether the data of one grid cell can predict the data of another grid cell. If the data of one grid cell can significantly improve the prediction accuracy of the data of adjacent grid cells after being added to the vector autoregressive model, it is considered that there is a causal relationship. Through spatial autocorrelation analysis methods, such as Moran's index, first, the multi-source heterogeneous data of each grid cell is used as a spatial variable, and the spatial autocorrelation coefficient of these variables between adjacent grid cells is calculated. By calculating Moran's index, the degree of spatial correlation of data between adjacent grid cells is judged.

[0097] Integrate the results of causal relationship analysis and spatial correlation analysis to form the spatio-temporal correlation relationship between adjacent grid cells. For example, comprehensively consider the strength of the causal relationship (measured by the coefficient size of Granger causality test) and the spatial autocorrelation coefficient to construct a spatio-temporal correlation relationship matrix, and the matrix elements represent the spatio-temporal correlation degree of different aspects between adjacent grid cells. Thus, it deeply reveals the interaction mechanism of traffic situations between adjacent grid cells, provides strong support for understanding the propagation law of traffic flow in the road network, and helps to more accurately grasp the overall operation state of the traffic road network.

[0098] As a quantitative index, spatio-temporal conduction entropy can comprehensively reflect the uncertainty and mutual influence degree of traffic situations between adjacent grid cells in the spatio-temporal dimension. Based on the entropy concept in information theory, combined with the spatio-temporal correlation relationship obtained before, spatio-temporal conduction entropy is defined and calculated. For example, using the maximum entropy principle, under the condition of satisfying the known spatio-temporal correlation relationship constraints, an optimization problem is solved to obtain the value of spatio-temporal conduction entropy. Specifically, let the spatio-temporal correlation relationship between adjacent grid cells be expressed as a set of constraint conditions, and by maximizing the entropy function while satisfying the constraint conditions, the obtained value is the spatio-temporal conduction entropy. In actual calculation, methods such as the Lagrange multiplier method combined with the gradient descent method can be used to solve this optimization problem. Thus, the complex spatio-temporal correlation relationship is quantified into a specific numerical value, which is convenient for subsequent numerical calculation and analysis, provides an objective and comparable basis for determining the basic field strength of each grid cell, helps to more accurately generate the situation field strength map, and thus intuitively displays the situation distribution of the traffic road network.

[0099] S3. Extract the situation feature tensor of the situation field strength map. The situation feature tensor includes situation layer features, event layer features, and deduction layer features. Visual encoding is performed on the situation feature tensor according to the basic field strength of each grid cell to determine the hierarchical rendering instruction.

[0100] Specifically, the extraction logic of the situation feature tensor includes:

[0101] Divide the situation field strength map into a situation layer, an event layer, and a deduction layer;

[0102] Configure multi-layer extraction units to parallelly extract the situation layer features, event layer features, and deduction layer features of the situation field strength map through the multi-layer extraction units;

[0103] Fuse the situation layer features, event layer features, and deduction layer features through an attention mechanism to obtain the situation feature tensor.

[0104] The traffic road network situation contains various different types of information. Through hierarchical division, features of different natures can be separately processed, which is convenient for more targeted extraction of the unique information of each layer, improving the accuracy and efficiency of feature extraction. The situation layer mainly reflects the normal operation situation of the traffic road network, such as traffic flow and speed distribution, etc. The event layer focuses on the impact of special events on traffic, such as traffic accidents and road construction, etc. The deduction layer is the prediction information of the future traffic situation based on the existing data. This hierarchical method can clearly sort out the complex traffic situation information and provide a basis for subsequent accurate feature extraction; in the data structure of the situation field strength map, different storage areas or data tags are defined to distinguish the data of each layer. For example, a three-dimensional array is used to store the situation field strength map data. The first dimension of the array represents the number of grid cells, the second dimension represents the feature dimension, such as the basic field strength and the spatio-temporal conduction entropy, etc., and the third dimension is used to distinguish the situation layer, the event layer, and the deduction layer. For the situation layer, the data related to the normal state of the traffic flow is stored in the corresponding dimension position. The event layer stores the attribute information related to the event, such as the event type, occurrence time, and influence range, etc. The deduction layer stores the data related to the future traffic situation predicted based on historical data, such as the predicted traffic flow change trend and the probability of congestion occurrence, etc. Through this data structure design, a clear hierarchical division of the situation field strength map is realized; thus, the complex traffic situation data is structured, and the information of each layer is independent and clear, which helps to adopt more appropriate algorithms according to the characteristics of different layers in the subsequent feature extraction process, improving the accuracy and efficiency of feature extraction.

[0105] Features of different layers have different properties and feature extraction requirements. Parallel extraction can make full use of distributed computing resources, accelerate the feature extraction speed, and improve the processing efficiency. Spatial features in the situation layer data are extracted through a convolutional neural network. For example, the situation layer data is scanned by convolutional kernels of different sizes to capture the distribution patterns and changing trends of traffic flow in space. Then, principal component analysis is used to reduce the dimensions of the features extracted by the convolutional neural network, remove redundant information, and retain the most representative features. For example, for the high-dimensional feature vector output by the CNN, the eigenvalues and eigenvectors are calculated using principal component analysis, and the principal components with a cumulative contribution rate reaching a certain threshold (such as 95%) are selected as the features of the situation layer.

[0106] The text information in the event layer (such as accident descriptions and construction notices, etc.) is preprocessed through natural language processing techniques, including word segmentation, part-of-speech tagging, and stop word removal, etc. Then, the preprocessed text data is converted into a word vector representation and input into a long short-term memory network. Through the long short-term memory network, a series of text data related to events is learned to extract event layer features, such as the type, impact degree, and duration of events, etc. The autoregressive integrated moving average model is used to model historical traffic situation data to predict future trends. Through continuous iterative training, features that can accurately reflect future traffic situation changes are extracted. For example, historical traffic flow data is input into the autoregressive integrated moving average model to obtain prediction results to extract the features of the deduction layer. Thus, the efficiency of feature extraction is greatly improved, the processing time is shortened, enabling the analysis of a large amount of traffic situation data in a relatively short time. At the same time, the customized extraction methods for different layer features improve the accuracy and pertinence of feature extraction, and can better mine the key information in each layer of data.

[0107] The features of different layers have different importance under different traffic scenarios and decision-making requirements. The attention mechanism can automatically learn the weights of the features of each layer, dynamically adjust the contribution of the features of each layer to the final result according to the actual situation, so as to more accurately reflect the comprehensive situation of the traffic road network. For example, when there is traffic congestion, the features of the event layer (such as traffic accident information) are more critical for decision-making, while under normal traffic conditions, the features of the situation layer (such as traffic flow distribution) are more important. The query vector, key vector, and value vector corresponding to each layer are obtained through the attention mechanism, and then the similarity score between the query vector and the key vector is calculated through dot product operation. After normalization, the attention weights of the features of each layer are obtained. Finally, the value vectors are weighted and summed according to the attention weights to obtain the fused situation feature tensor; thus, it can adaptively adjust the weights of the features of each layer, highlight the key features, and improve the ability to accurately describe and analyze the traffic situation. Compared with simple feature splicing or average fusion methods, the attention mechanism can better capture the complex relationships between the features of different layers and generate a more representative situation feature tensor, providing more reliable data support for subsequent visual coding and determination of hierarchical rendering instructions.

[0108] Specifically, the determination logic of the hierarchical rendering instructions includes:

[0109] Analyze the mapping relationship between the basic field strength of each grid cell and the situation feature tensor through a neural network;

[0110] Perform visual coding on the situation layer features, event layer features, and deduction layer features respectively, and adjust the parameters of the visual coding in combination with the basic field strength of each grid cell;

[0111] Generate the rendering instructions for the situation layer, event layer, and deduction layer according to the results of the visual coding and the mapping relationship between the basic field strength of each grid cell and the situation feature tensor to determine the hierarchical rendering instructions. (The rendering instructions include the visual element type, attribute value, and position in the traffic road network)

[0112] The basic field strength reflects the relative importance of each grid cell in the traffic road network and the intensity of the traffic situation, while the situation feature tensor contains the comprehensive situation information of the traffic road network. Analyzing the mapping relationship between them can help understand how the basic field strength affects and correlates with the overall traffic situation features, providing a basis for subsequent visual encoding of the situation feature tensor based on the basic field strength. By using a multi-layer perceptron neural network and combining the basic field strength of each grid cell, a predicted value corresponding to the dimension of the situation feature tensor is obtained. During the training process, a large number of basic field strengths of grid cells and their corresponding real situation feature tensor data are used as training samples. Through the backpropagation algorithm, the weights and biases of the multi-layer perceptron neural network are continuously adjusted, enabling the multi-layer perceptron neural network to learn the mapping relationship between the basic field strength and the situation feature tensor. Thus, a quantitative relationship between the basic field strength and the situation feature tensor is established, providing a scientific basis for subsequent visual encoding and determination of hierarchical rendering instructions based on the basic field strength. Through this mapping relationship, the key information of the basic field strength can be effectively integrated into the visualization process of the traffic situation, improving the accuracy of the visualization results and the guiding value for users.

[0113] Visual encoding can transform abstract traffic situation features into intuitive visual elements for easy understanding and observation by users. Adjusting visual encoding parameters in combination with the basic field strength can highlight areas with higher basic field strength, making the display of the traffic situation more focused and hierarchical. For situation layer features such as traffic flow, color mapping is used for visual encoding. A color mapping table is defined. According to the range of traffic flow magnitudes, different flow values are mapped to different colors. For example, green is set to represent low flow, yellow for medium flow, and red for high flow. Then, based on the situation layer features (i.e., traffic flow) of each grid cell, the corresponding color is selected from the color mapping table, and the brightness of the color is adjusted in combination with the basic field strength. For grid cells with a larger basic field strength, the color brightness is appropriately increased to highlight the display.

[0114] For the event layer features, an icon library is used for encoding. Different icons are defined for different types of events (such as traffic accidents, road construction, and traffic control). According to the event layer feature information of each grid cell, the corresponding icon is selected from the icon library and displayed at the position of the grid cell. The size of the icon is adjusted in combination with the basic field strength. For grid cells with a larger basic field strength, the icon size is appropriately increased. For the deduction layer features, visual encoding is carried out in the form of dynamic graphics. For example, the thickness and color change of lines are used to represent the change trend of future traffic flow. The line becoming thicker indicates an increase in flow, and the color becoming darker indicates a larger increase amplitude. The display speed of the dynamic graphics is adjusted in combination with the basic field strength. For grid cells with a larger basic field strength, the display speed of the dynamic graphics is appropriately increased to highlight the future situation changes in this area. The speed adjustment is achieved by setting the frame rate of the animation. Thus, the abstract traffic situation features are transformed into intuitive and easy-to-understand visual elements, and the key areas can be highlighted according to the basic field strength, improving the effect of traffic situation visualization and the information transmission efficiency. Users can quickly understand the overall situation of the traffic road network and the situation of key areas by observing the visualization results, providing intuitive support for traffic decision-making.

[0115] Rendering instructions are to convert the visually encoded information into commands that the computer can understand and execute, for accurately drawing the traffic situation on the visualization interface. According to the previous visual encoding results and mapping relationships, rendering instructions containing information such as the type of visual element, attribute values, and positions in the traffic road network are generated. For the situation layer, the generated rendering instructions include the color value corresponding to each grid cell (determined according to visual encoding), the color brightness adjustment value (determined in combination with the basic field strength), and the geographical coordinate position of the grid cell in the traffic road network. For the event layer, the rendering instructions include the icon type corresponding to each grid cell (determined according to the event type), the icon size adjustment value (determined in combination with the basic field strength), and the coordinate position. For the deduction layer, the rendering instructions include the parameters of the dynamic graphics (such as line thickness, color change, and frame rate) and the coordinate position. These rendering instructions for the situation layer, event layer, and deduction layer are integrated to form complete hierarchical rendering instructions for hierarchical rendering of the traffic road network situation on the visualization software or platform. The generated hierarchical rendering instructions can accurately guide the computer to draw graphics that conform to the actual traffic situation on the visualization interface, highlighting key areas and making the display of the traffic road network situation clearer and more intuitive. Users can quickly obtain comprehensive and key traffic information through the visualization interface, providing strong support for traffic management and decision-making.

[0116] S4. Obtain eye movement tracking data to dynamically adjust the basic field strength of each grid cell, and semantically describe and position annotate the hierarchical rendering instructions through natural language processing and annotation positioning algorithms. Conduct multi-layer situation deduction of the traffic road network according to the hierarchical rendering instructions, and dynamically adjust the grid granularity according to the error of the multi-layer situation deduction.

[0117] Specifically, the adjustment logic for the basic field strength of each grid cell includes:

[0118] Obtain the user's eye movement tracking data in real time. The eye movement tracking data includes the fixation point position, fixation time, and saccade path, and map the eye movement tracking data to the grid cells.

[0119] Identify the grid cells that the user is interested in through a clustering algorithm to obtain the degree of attention for each grid cell.

[0120] Determine the adjustment direction and adjustment amplitude of the basic field strength for each grid cell according to the degree of attention of each grid cell.

[0121] Obtain the adjusted basic field strength for each grid cell based on the adjustment direction and adjustment amplitude of the basic field strength for each grid cell and the current basic field strength of each grid cell.

[0122] The user's eye movement behavior can intuitively reflect their degree of attention to different areas of the traffic road network. By obtaining this data and mapping it to the corresponding grid cells, the interest points of the user in the traffic situation of each area can be understood, providing a basis for subsequently adjusting the basic field strength according to the user's attention degree, making the displayed traffic situation more in line with the user's needs; capturing the movement of the user's eyes in real time through an infrared camera, using image processing algorithms to analyze information such as the pupil position and the angle of eye rotation, so as to determine the fixation point position. The fixation time is obtained by recording the duration of the user's stay at a certain fixation point, and the saccade path is determined by tracking the continuous change of the fixation point. Connect the infrared camera to the terminal that displays the traffic road network situation, establish a grid coordinate system corresponding to the traffic road network topology on the display interface, and map it to the grid cells of the traffic road network according to the coordinates of the fixation point on the screen through a coordinate conversion algorithm. For example, given parameters such as the screen resolution, the position of the display interface on the screen, and the scaling ratio of the traffic road network topology on the display interface, calculate the grid cell number corresponding to the fixation point through a formula; thus, the association between the user's subjective attention information and the objective traffic road network area is realized, providing a data basis for personalized adjustment of the traffic situation display, being able to highlight the traffic situation of relevant areas according to the user's focus of attention, and improving the efficiency of the user to obtain key information.

[0123] The directly obtained eye movement tracking data is relatively scattered, making it difficult to intuitively judge the degree of attention of users to each grid cell. The clustering algorithm can group grid cells with similar eye movement characteristics (such as dense fixation points and long fixation times, etc.) into one category, thereby identifying the areas that users focus on, and quantifying the degree of attention of each grid cell. The DBSCAN density clustering algorithm is selected to divide data points into core points, boundary points, and noise points. For each grid cell, calculate the number of other grid cells contained within its neighborhood radius area (i.e., density). If the density of a certain grid cell is greater than or equal to the minimum number of points, it is marked as a core point. If the density of a certain grid cell is less than the minimum number of points but within the neighborhood radius area of a certain core point, it is marked as a boundary point. Other grid cells are marked as noise points. Clusters are formed by connecting core points and boundary points within their neighborhood radius areas. The degree of attention can be represented by the ratio of the number of core points in the cluster where the grid cell is located to the total number of core points. For example, calculate the total number of core points in all clusters and the number of core points in the cluster where a certain grid cell is located. Then the degree of attention of this grid cell is the ratio of the number of core points in the cluster where the grid cell is located to the total number of core points in all clusters. Thus, it can automatically and accurately identify the grid cells that users are interested in from a large amount of scattered eye movement data, and quantify the degree of attention, providing a scientific basis for subsequent targeted adjustment of the basic field strength. Compared with manual judgment or simple statistical methods, the clustering algorithm can more comprehensively and objectively reflect the attention pattern of users.

[0124] The areas that users are concerned about are usually the focus of traffic situation analysis and need to be highlighted. By determining the adjustment direction and amplitude of the basic field strength according to the degree of concern, the displayed traffic situation can be made more in line with user needs, highlighting the areas that users care about, while not affecting the basic display of other areas. The adjustment direction determines whether to increase or decrease the basic field strength, and the adjustment amplitude determines the degree of increase or decrease. The combination of the two can achieve fine control of the traffic situation display; establish an adjustment rule table, divide the degree of concern into different intervals, and each interval corresponds to different adjustment directions and amplitudes. For example, when the degree of concern is greater than or equal to 0.8, the adjustment direction is to increase the basic field strength, and the adjustment amplitude is to increase by 30% on the basis of the original basic field strength; when the degree of concern is greater than or equal to 0.5 and less than 0.8, the adjustment direction is to increase the basic field strength, and the adjustment amplitude is to increase by 15%; when the degree of concern is less than 0.5, the adjustment direction is to keep the basic field strength unchanged or appropriately weaken it according to the actual situation, such as weakening by 5%. In actual applications, these interval ranges and corresponding adjustment parameters can be flexibly adjusted according to user feedback and the needs of traffic situation analysis. By querying the adjustment rule table and according to the degree of concern of each grid cell, determine the adjustment direction and amplitude of its basic field strength; thus realizing personalized and precise adjustment of the basic field strength according to the user's degree of concern, making the traffic situation display more targeted and practical, enabling users to obtain traffic information in their concerned areas more quickly and clearly, and improving the user experience and information transmission efficiency of traffic situation data display.

[0125] Through the previous steps, the adjustment direction and amplitude are determined, and combined with the current basic field strength for calculation, the finally adjusted basic field strength can be obtained, thus realizing real-time update of the traffic road network situation display, making the display result more in line with user needs and the actual traffic situation; obtain the adjusted basic field strength according to the determined adjustment direction and amplitude. In actual calculation, by traversing each grid cell, according to its corresponding adjustment direction and amplitude, obtain the adjusted basic field strength and update the corresponding value in the basic field strength data storage structure; thus completing the adjustment process of the basic field strength, enabling the traffic road network situation display to be optimized in real time according to the user's concern, highlighting the areas that users care about, providing more valuable traffic information for users, and at the same time the adjusted basic field strength data provides input data more in line with user needs for subsequent multi-layer situation deduction and other related analyses.

[0126] Specifically, the adjustment logic of the grid granularity includes:

[0127] In the process of multi-layer situation deduction of the traffic road network according to the hierarchical rendering instruction, the mean square error is used to record in real time the error size and error change trend between the deduction results of the situation layer, event layer and deduction layer and the actual traffic conditions;

[0128] Using the fuzzy logic inference method, determine whether it is necessary to adjust the grid granularity and the adjustment direction of the grid granularity according to the error magnitude, the error change trend, and the actual traffic conditions;

[0129] Adjust the grid granularity according to the adjustment direction of the grid granularity, and conduct multi-layer situation deduction again. Compare the error magnitudes before and after adjusting the grid granularity to determine whether to readjust the grid granularity.

[0130] The purpose of multi-layer situation deduction is to predict the future situation of the traffic road network. Since the actual traffic conditions are constantly changing, by recording the errors between the deduction results and the actual situation in real time, the accuracy and reliability of the current situation deduction model can be evaluated. The error magnitude reflects the deviation degree between the current deduction prediction value and the true value, and the error change trend can reflect the change of the deduction prediction performance over time. These information are crucial for determining whether to adjust the grid granularity and how to adjust it, because the selection of the grid granularity will affect the ability of the situation deduction to capture the details of the traffic road network and the calculation efficiency.

[0131] During the multi-layer situation deduction process, for the situation layer, obtain the actual traffic flow data in real time, such as traffic volume and vehicle speed, and compare them with the corresponding data in the deduction results. Similarly, for the event layer, compare the event information that actually occurs (such as event type, occurrence time, and location, etc.) with the deduction results. Calculate the mean square error through a certain quantization method (such as encoding the event type as a numerical value and calculating the errors of the event occurrence time and location). For the deduction layer, compare the predicted future traffic situation based on the current model with the actually observed subsequent traffic situation to calculate the mean square error. Use the moving average method to smooth these mean square errors to highlight the error change trend. By plotting the curve of the mean square error changing with time, visually display the error magnitude and change trend; thus, it can evaluate the performance of the multi-layer situation deduction model in real time and quantitatively, providing an objective basis for subsequent adjustment of the grid granularity. Through the analysis of the error magnitude and change trend, it can be found in time that the deduction prediction is inaccurate, so as to adjust the grid granularity targeted and improve the accuracy of the situation deduction.

[0132] The complexity and dynamics of the traffic road network make it difficult to determine whether to adjust the grid granularity and how to adjust it through simple rules or fixed thresholds. Fuzzy logic inference can handle uncertain and fuzzy information, comprehensively consider multiple factors such as the error magnitude, the error change trend, and the actual traffic conditions, and make decisions that are more in line with the actual situation. For example, when the error is large and shows an upward trend, and the actual traffic conditions show that the traffic flow changes complexly, it is necessary to reduce the grid granularity to improve the accuracy of the situation deduction. When the error is small and stable, and the actual traffic conditions are relatively stable, there is no need to adjust the grid granularity or appropriately increase the grid granularity to improve the calculation efficiency.

[0133] Construct a fuzzy logic inference unit, which consists of four parts: fuzzification, fuzzy rule base, fuzzy inference, and defuzzification. Fuzzify the input variables such as the magnitude of the error, the trend of error change, and the actual traffic conditions. For example, divide the magnitude of the error into three fuzzy sets: "small", "medium", and "large". Map the specific mean square error value to the membership degree of the corresponding fuzzy set through the membership function. Similarly, divide the trend of error change into "decreasing", "stable", and "increasing", and divide the actual traffic conditions into fuzzy sets such as "simple", "general", and "complex", and establish the corresponding membership functions.

[0134] According to the experience of traffic domain experts and actual situations, formulate a series of fuzzy rules. For example, Rule 1: If the magnitude of the error is "large", the trend of error change is "increasing", and the actual traffic condition is "complex", then the adjustment direction of the grid granularity is "decrease"; Rule 2: If the magnitude of the error is "small", the trend of error change is "stable", and the actual traffic condition is "simple", then the adjustment direction of the grid granularity is "unchanged". Through a large number of experiments and data analysis, continuously optimize and improve the fuzzy rule base.

[0135] Through the Mamdani inference method, based on the fuzzification results of the input variables and the fuzzy rule base, conduct fuzzy inference. For example, for the input at a certain moment, according to the membership degree values after fuzzification, activate the corresponding fuzzy rules, and obtain the fuzzy set of the inference result through the minimum operation; defuzzify the result obtained from the fuzzy inference to obtain a specific grid granularity adjustment decision. Common defuzzification methods include the centroid method, that is, calculate the centroid position of the fuzzy set and convert it into a specific adjustment direction, such as "increase", "decrease", "unchanged"; thus, it can comprehensively consider multiple complex factors, make a more reasonable and intelligent grid granularity adjustment decision, adapt to the dynamic changes and uncertainties of the traffic road network. Compared with the traditional method based on fixed thresholds or simple rules, fuzzy logic inference can more flexibly handle different traffic scenarios and improve the adaptability and accuracy of the situation deduction model.

[0136] After making the actual adjustment according to the grid granularity adjustment direction determined by the fuzzy logic inference, it is necessary to conduct multi-layer situation deduction again. By comparing the magnitude of the error before and after the adjustment, evaluate the effect of the adjustment. If the error decreases significantly after the adjustment, it indicates that the adjustment direction is correct and the grid granularity is more suitable for the current traffic road network situation. If the error does not improve or even increases, it is necessary to re-consider the adjustment strategy, re-adjust the grid granularity or check other parameter settings of the situation deduction model.

[0137] If the grid granularity adjustment direction is "increase", in the geographic information system environment, the original grid cells are merged. For example, four adjacent small grid cells are merged into one large grid cell, and relevant parameters such as the basic field strength and spatio-temporal conduction entropy of the merged grid cell are recalculated. If the adjustment direction is "decrease", the original grid cells are subdivided. For example, one large grid cell is divided into four small grid cells, and the relevant parameters of the subdivided grid cells are also recalculated. After completing the adjustment of the grid granularity and parameter calculation, the multi-layer situation deduction is run again, and the mean square error between the deduced result after adjustment and the actual traffic condition is recorded. The mean square error after adjustment is compared with the mean square error before adjustment. If the mean square error after adjustment is less than the mean square error before adjustment and meets certain threshold conditions, such as the error reduction amplitude is greater than 10%, it is considered that the adjustment is effective and the current grid granularity is maintained. If this condition is not met, return to the fuzzy logic reasoning step to re-evaluate whether the grid granularity needs to be adjusted and the adjustment direction, or optimize other parameters of the situation deduction model.

[0138] Through actual adjustment and effect evaluation, the grid granularity can be continuously optimized, and the accuracy and adaptability of the multi-layer situation deduction can be improved. This iterative optimization process can make the traffic road network situation data display method better adapt to different traffic scenarios and changes, providing more reliable support for traffic management and decision-making. After the grid granularity adjustment and effect evaluation, the determined final grid granularity will affect the subsequent entire traffic road network situation data display process. A suitable grid granularity can make steps such as the generation of the situation field strength map, the extraction of situation feature tensors, the determination of hierarchical rendering instructions, and the multi-layer situation deduction more accurate and efficient, thereby improving the quality and practicality of the traffic road network situation data display.

[0139] Embodiment 2

[0140] As Figure 3 shown, the embodiment of the present application provides a system module diagram of a traffic road network situation data display system. The system includes a topology generation module, a map generation module, a hierarchical rendering module, and a feedback adjustment module.

[0141] The topology generation module is used to obtain multi-source heterogeneous data including traffic flow data, radar trajectory data, and situation propagation data in the traffic road network, perform real-time interpolation on the missing multi-source heterogeneous data through a generative adversarial network, and perform spatio-temporal synchronization on the interpolated multi-source heterogeneous data to generate a traffic road network topology.

[0142] The map generation module is used to decompose the traffic road network topology into grid cells based on the grid granularity, calculate the spatio-temporal conduction entropy between adjacent grid cells, and determine the basic field strength of each grid cell according to the spatio-temporal conduction entropy between adjacent grid cells to generate a situation field strength map of the traffic road network.

[0143] The hierarchical rendering module is used to extract the situation feature tensor of the situation field strength map. The situation feature tensor includes situation layer features, event layer features, and deduction layer features. Visual coding is performed on the situation feature tensor according to the basic field strength of each grid cell to determine the hierarchical rendering instruction;

[0144] The feedback adjustment module is used to obtain eye movement tracking data to dynamically adjust the basic field strength of each grid cell, and perform semantic description and position annotation on the hierarchical rendering instruction through natural language processing and annotation positioning algorithms. Multilayer situation deduction of the traffic road network is carried out according to the hierarchical rendering instruction, and the grid granularity is dynamically adjusted according to the error of the multilayer situation deduction.

[0145] For the principle of the above system, please refer to the implementation steps of the above method, which will not be elaborated here one by one.

Claims

1. A method for displaying traffic network situation data, characterized in that: include: Acquire multi-source heterogeneous data in the traffic network, including traffic flow data, radar trajectory data, and event propagation data, interpolate the missing multi-source heterogeneous data in real time through a generative adversarial network, and synchronize the interpolated multi-source heterogeneous data in time and space to generate the traffic network topology; Based on the grid granularity, the traffic network topology is decomposed into grid units, the spatiotemporal conduction entropy between adjacent grid units is calculated, and the basic field strength of each grid unit is determined according to the spatiotemporal conduction entropy between adjacent grid units to generate the situation field strength map of the traffic network; Extracting the situation feature tensor of the situation field strength map, the situation feature tensor includes situation layer features, event layer features and deduction layer features, and visually encoding the situation feature tensor according to the basic field strength of each grid unit to determine the layered rendering instructions; Obtain eye tracking data to dynamically adjust the basic field strength of each grid unit, and use natural language processing and annotation positioning algorithms to semantically describe and position the layered rendering instructions, perform multi-layer situation deduction of the traffic network according to the layered rendering instructions, and dynamically adjust the grid granularity according to the error of the multi-layer situation deduction; the error size and error change trend between the deduction results of the situation layer, event layer and deduction layer and the actual traffic conditions are recorded in real time through the mean square error; Through the fuzzy logic reasoning method, according to the error size, error change trend and actual traffic conditions, determine whether the grid granularity needs to be adjusted and the adjustment direction of the grid granularity; adjust the grid granularity according to the adjustment direction of the grid granularity, and perform multi-layer situation deduction again to compare the error size before and after adjusting the grid granularity to determine whether the grid granularity needs to be readjusted.

2. A method for displaying traffic network situation data according to claim 1, characterized in that: The generation logic of the traffic network topology includes: Acquire multi-source heterogeneous data including traffic flow data, radar trajectory data and event propagation data from the traffic network; Real-time interpolation of missing multi-source heterogeneous data through generative adversarial networks; The interpolated multi-source heterogeneous data are synchronized in time and space according to the constructed spatial reference points and time reference points; The multi-source heterogeneous data that have been synchronized in time and space are used to determine the connection relationship of nodes and the weight of edges in the transportation network through a topological algorithm to generate the transportation network topology.

3. A method for displaying traffic network situation data according to claim 2, characterized in that: The generation logic of the situation field strength map includes: Based on the grid granularity, the traffic network topology is decomposed into grid units, and the spatiotemporal conduction entropy between adjacent grid units is calculated; The space-time conduction entropy between adjacent grid cells is used as a weight to determine the basic field strength of each grid cell by weighted average. The basic field strength and space-time conduction entropy of each grid unit are integrated to generate a situation field strength map, which is then visualized.

4. A method for displaying traffic network situation data as claimed in claim 3, characterized in that: The decomposition sub-logic of the grid unit includes: The convolutional neural network is used to extract the features of the traffic network topology and identify the traffic network type; Determine the grid granularity according to the type of traffic network; The traffic network topology is decomposed into grid cells based on the grid granularity, and the boundary of each grid cell is determined according to its position in the traffic network.

5. A method for displaying traffic network situation data as claimed in claim 4, characterized in that: The calculation sub-logic of the spatiotemporal conduction entropy between adjacent grid cells includes: Extract multi-source heterogeneous data for each grid cell from the traffic network topology; Analyze the causal relationship and spatial correlation of multi-source heterogeneous data between adjacent grid cells to obtain spatiotemporal correlation; The spatiotemporal conduction entropy between adjacent grid cells is calculated based on the spatiotemporal correlation relationship.

6. A method for displaying traffic network situation data according to claim 5, characterized in that: The extraction logic of the situation feature tensor includes: The situation field strength map is divided into situation layer, event layer and deduction layer; A multi-layer extraction unit is configured to extract situation layer features, event layer features and deduction layer features of the situation field strength map in parallel through the multi-layer extraction unit; The situation layer features, event layer features and inference layer features are fused through the attention mechanism to obtain the situation feature tensor.

7. A method for displaying traffic network situation data according to claim 6, characterized in that: The determination logic of the layered rendering instruction includes: The mapping relationship between the basic field strength and situational characteristic tensor of each grid unit is analyzed through a neural network; Visually encode the situation layer features, event layer features and deduction layer features respectively, and adjust the parameters of the visual encoding in combination with the basic field strength of each grid unit; According to the results of visual encoding and the mapping relationship between the basic field strength of each grid unit and the situation feature tensor, rendering instructions for the situation layer, event layer and deduction layer are generated to determine the hierarchical rendering instructions.

8. A method for displaying traffic network situation data according to claim 7, characterized in that: The adjustment logic of the basic field strength of each grid unit includes: Obtain the user's eye tracking data in real time, including gaze point location, gaze time, and scan path, and map the eye tracking data to grid cells; The grid units that users pay attention to are identified through clustering algorithms, and the degree of attention of each grid unit is obtained; Determine the adjustment direction and adjustment amplitude of the basic field strength of each grid unit according to the attention degree of each grid unit; According to the adjustment direction and adjustment amplitude of the basic field strength of each grid unit and the current basic field strength of each grid unit, the adjusted basic field strength of each grid unit is obtained.

9. A traffic network situation data display system, characterized in that: include: Topology generation module, graph generation module, layered rendering module and feedback adjustment module; The topology generation module is used to obtain multi-source heterogeneous data in the traffic network, including traffic flow data, radar trajectory data, and event propagation data. The missing multi-source heterogeneous data is interpolated in real time through the adversarial generative network, and the interpolated multi-source heterogeneous data is synchronized in time and space to generate the traffic network topology. The map generation module is used to decompose the traffic network topology into grid units based on the grid granularity, calculate the spatiotemporal conduction entropy between adjacent grid units, and determine the basic field strength of each grid unit according to the spatiotemporal conduction entropy between adjacent grid units to generate a situational field strength map of the traffic network; The hierarchical rendering module is used to extract the situation feature tensor of the situation field strength map. The situation feature tensor includes situation layer features, event layer features and deduction layer features. The situation feature tensor is visually encoded according to the basic field strength of each grid unit to determine the hierarchical rendering instructions. The feedback adjustment module is used to obtain eye tracking data to dynamically adjust the basic field strength of each grid unit, and to semantically describe and position the layered rendering instructions through natural language processing and annotation positioning algorithms, to perform multi-layer situation deduction of the traffic network according to the layered rendering instructions, and to dynamically adjust the grid granularity according to the error of the multi-layer situation deduction; wherein the error size and error change trend between the deduction results of the situation layer, event layer and deduction layer and the actual traffic conditions are recorded in real time through the mean square error; Through the fuzzy logic reasoning method, according to the error size, error change trend and actual traffic conditions, determine whether the grid granularity needs to be adjusted and the adjustment direction of the grid granularity; adjust the grid granularity according to the adjustment direction of the grid granularity, and perform multi-layer situation deduction again to compare the error size before and after adjusting the grid granularity to determine whether the grid granularity needs to be readjusted.

Images