Urban governance data fusion analysis method and system

Abstract

The invention relates to the technical field of data fusion analysis, in particular to an urban governance data fusion analysis method and system. The method comprises the following steps: acquiring movement track data of pedestrians in a city by using a sensor, and acquiring position information data of vehicles in the city by using GPS equipment; fusing the motion trail data of the pedestrian and the position information data of the vehicle through a weighted average method; extracting tracks of pedestrians and vehicles from the fused data, and processing by using a dynamic time warping algorithm; the method comprises the following steps: modeling a space-time mode of pedestrian and vehicle tracks based on a support vector machine model, identifying road traffic anomalies, and introducing a traffic jam function in a modeling process for traffic jam; according to the design of the invention, the dynamic time warping (DTW) algorithm is introduced and the Euclidean distance calculation method is optimized, so that the condition that pedestrians and vehicles wait in front of the traffic lights can be better processed, and unnecessary distance punishment caused by a static state is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of data fusion analysis, and in particular to a method and system for urban governance data fusion analysis. Background Art

[0002] With the growth of urban population and the increase of the number of vehicles, traffic congestion has become more serious, resulting in a lot of time and resource waste, and also increasing the risk of traffic accidents. Especially during peak hours, traffic light control, pedestrian crossing, vehicle speed, parking behavior, etc. may affect traffic flow; in the urban road environment shared by people and vehicles, the interaction between pedestrians and vehicles may cause accidents. Pedestrians running red lights, vehicles speeding, and not stopping according to regulations not only threaten the safety of pedestrians, but also affect traffic order. Therefore, real-time monitoring and prediction of these behaviors and taking appropriate management measures have become the key to improving urban traffic safety; existing urban governance methods often rely on manual setting of traffic lights, manual inspections, and fixed traffic rules, but this governance method has certain limitations and cannot be dynamically adjusted according to changes in traffic flow or emergencies in real time. The data collected by various urban sensors (such as traffic monitoring cameras, GPS devices, mobile terminals, etc.) are often scattered in different systems, lacking effective fusion analysis, resulting in data islands. Therefore, a method and system for urban governance data fusion analysis is provided. Summary of the invention

[0003] The purpose of the present invention is to provide a method and system for urban governance data fusion analysis to solve the problems raised in the above background technology that traditional urban governance relies on manual intervention, information islands, and safety hazards between pedestrians and vehicles.

[0004] To achieve the above object, the present invention provides a method for fusion analysis of urban governance data, comprising: S1. Use sensors to obtain the movement trajectory data of pedestrians in the city, and use GPS devices to obtain the location information data of vehicles in the city; S2, merging the pedestrian's motion trajectory data and the vehicle's location information data through a weighted average method; S3, extract the trajectories of pedestrians and vehicles from the fused data and process them using the dynamic time warping algorithm; S4. Based on the support vector machine model, the spatiotemporal patterns of pedestrian and vehicle trajectories are modeled to identify road traffic anomalies, and for traffic congestion, a traffic congestion function is introduced in the modeling process.

[0005] As a further improvement of the technical solution, in S1, the pedestrian's motion trajectory data includes: position coordinates , Timestamp .

[0006] As a further improvement of the technical solution, in S2, the pedestrian's motion trajectory data and the vehicle's position information data are fused by weighted average method, including the following steps: S2.1, preprocessing the motion trajectory data of pedestrians and the location information data of vehicles; S2.2, converting the motion trajectory data of all pedestrians and the position information data of vehicles into a unified coordinate system; S2.3, using interpolation method to time align the motion trajectory data of the pedestrian and the position information data of the vehicle; S2.4. The coordinates of pedestrians and vehicles are fused by weighted averaging to obtain the fused coordinates: ; ; in, Represents the weight coefficient of pedestrian motion trajectory data, Represents the weight coefficient of vehicle location information data, Represents the horizontal coordinate of the fused data. Represents the vertical coordinate of the fused data. represents the position coordinates of the vehicle, Indicates the time point that needs to be interpolated.

[0007] As a further improvement of the technical solution, in S3, the trajectories of pedestrians and vehicles are extracted from the fused data and processed using a dynamic time warping algorithm, including the following steps: S3.1, dividing the fused continuous data stream into independent trajectory segments according to the time period t1 to be analyzed, and preprocessing the trajectory segments; S3.2. The distance between trajectory points is calculated by the Euclidean distance method. The Euclidean distance method is optimized for the waiting time of traffic lights on urban roads. The Euclidean distance method is further optimized by considering the impact of traffic flow on the waiting time of traffic lights. S3.3, calculating the cumulative cost matrix by recursion based on the distance between trajectory points; S3.4. Starting from the lower right corner of the cumulative cost matrix, find the optimal alignment path by tracing back the path with the minimum cost; S3.5. Align the corresponding points in the pedestrian trajectory and the vehicle trajectory by tracing the optimal alignment path, and obtain the reconstructed trajectory of the pedestrian trajectory and the vehicle trajectory in time and space.

[0008] As a further improvement of the technical solution, in S3.3, the distance between the trajectory points is calculated by the Euclidean distance method as follows: ; in, represents the pedestrian trajectory point, represents the vehicle trajectory point, represents the distance between the pedestrian trajectory point and the vehicle trajectory point, represents the first The timestamp corresponding to each trajectory point; Represents the vehicle trajectory data The timestamp corresponding to each trajectory point; The Euclidean distance method is optimized for the waiting time of traffic lights on urban roads: ; in, It represents the distance between the trajectory points of pedestrians and vehicles after considering the waiting time of traffic lights on urban roads; Indicates the degree of influence of pedestrian waiting time on the total distance; Indicates the degree of influence of vehicle waiting time on the total distance; represents the waiting time of pedestrians due to traffic lights; Indicates the waiting time of vehicles due to traffic lights; Considering the impact of traffic flow on the waiting time of traffic lights, the Euclidean distance method is further optimized: ; in, represents the distance between the pedestrian trajectory point and the vehicle trajectory point after further optimization; represents a function reflecting the traffic flow situation; Represents the weight coefficient of traffic flow.

[0009] As a further improvement of the technical solution, in S3.3, the cumulative cost matrix is ​​calculated by recursion as follows: ; in, represents the minimum cumulative cost; Indicates that from the sequence point to another sequence The cumulative cost of each point; Indicates that from the sequence point to another sequence The cumulative cost of each point; Indicates that from the sequence point to another sequence The cumulative cost of each point.

[0010] As a further improvement of the technical solution, in S4, identifying road traffic anomalies by modeling the spatiotemporal patterns of pedestrian and vehicle trajectories based on a support vector machine model includes the following steps: S4.1, extract and analyze the interaction features between pedestrians and vehicles, extract spatiotemporal features from the fused pedestrian and vehicle trajectory data, and preprocess the extracted spatiotemporal features and interaction features; S4.2, based on the preprocessed spatiotemporal features and interaction features, select features related to identifying abnormal behaviors; S4.3, taking the spatiotemporal features and interaction features of each trajectory segment as sample input, constructing a training dataset and a test dataset; S4.4, select the kernel function to fit the data, optimize the kernel function for traffic congestion, and use the support vector machine model to classify abnormal behaviors; S4.5. Use the trained support vector machine model to determine whether the new trajectory data is abnormal.

[0011] As a further improvement of the technical solution, in S4.4, the kernel function is: ; in, Represents a given input point and The similarity calculated between Represents the kernel width, which controls the smoothness of the kernel function; The feature vector representing the pedestrian trajectory data point; The feature vector representing the vehicle trajectory data point; Optimize the kernel function for traffic congestion: ; ; in, represents the similarity calculated after optimization for traffic congestion; represents the weight of controlling the impact of traffic congestion on similarity; represents the traffic congestion function; represents the weight coefficient for controlling speed difference; Indicates the weight coefficient for controlling the state of the traffic light; Represents the weight coefficient for controlling traffic flow; Indicates the speed difference between pedestrians and vehicles; Represents the influence function of traffic lights on similarity; Represents the difference in traffic volume.

[0012] As a further improvement of the technical solution, in S4.5, the support vector machine model is: ; in, Represents the decision function output of the support vector machine model; represents the number of support vectors; represents the Lagrange multiplier; Indicates The category labels of training samples; Represents the index of the training sample; Represents the bias term.

[0013] On the other hand, the present invention provides an urban governance data fusion and analysis system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any one of the urban governance data fusion and analysis methods described above.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. The urban governance data fusion analysis method and system, by introducing the dynamic time warping (DTW) algorithm and optimizing the Euclidean distance calculation method, can better handle the situation of pedestrians and vehicles waiting at traffic lights, avoiding unnecessary distance penalties caused by static state. This not only improves the temporal and spatial alignment accuracy of pedestrian and vehicle trajectories, but also enhances the accuracy of abnormal behavior recognition, such as illegal red light running, abnormal parking, etc., thus helping to improve the efficiency and safety of urban traffic management.

[0015] 2. In the urban governance data fusion analysis method and system, by optimizing the kernel function of the support vector machine (SVM) model to take into account traffic congestion factors, this method can more accurately simulate the spatial and temporal patterns of pedestrians and vehicles under actual traffic conditions. This improvement enables the model to maintain high classification performance under different traffic flow conditions, and can respond to real-time changing traffic conditions more quickly and accurately, providing strong data support for urban governance and helping decision makers to formulate more scientific and reasonable traffic management policies. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 The figure is a flow chart of the overall method of the present invention. DETAILED DESCRIPTION

[0017] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0018] Example 1: Please refer to Figure 1 As shown, this embodiment provides a method for fusion analysis of urban governance data, including the following steps: S1. Use sensors to obtain the movement trajectory data of pedestrians in the city, and use GPS devices to obtain the location information data of vehicles in the city; In this embodiment, the motion trajectory data of the pedestrian includes: position coordinates , Timestamp .

[0019] S2, merging the pedestrian's motion trajectory data and the vehicle's location information data through a weighted average method; In this embodiment, the weighted average method is a calculation method that calculates the average by assigning different weights to different data points, so that some data points have a greater impact on the results than other data points; pedestrians and vehicles have different behavior patterns in urban environments, and they are affected by different factors. For example, pedestrians may be more flexible and can change directions in a short period of time; while vehicles are more restricted by traffic regulations, road conditions, etc. By fusing these two different types of data through the weighted average method, the advantages of both can be combined to reduce the errors or uncertainties that may be caused by a single data source, thereby obtaining more accurate and reliable comprehensive trajectory information; the data collection frequencies of pedestrians and vehicles may be different, and their coordinate systems may also be inconsistent (such as geographic coordinate system and plane rectangular coordinate system). Through the weighted average method, it is possible to ensure that the timestamps of pedestrian motion trajectory data and vehicle location information data are consistent or synchronized, and convert these data into a unified coordinate system. This helps to achieve precise alignment of pedestrian and vehicle trajectories in time and space, providing a solid foundation for subsequent analysis; The pedestrian's motion trajectory data and the vehicle's location information data are fused by weighted average method, including the following steps: S2.1. Preprocess the pedestrian's motion trajectory data and the vehicle's location information data to ensure that the timestamps of the pedestrian's motion trajectory data (including location coordinates and timestamps) and the vehicle's location information data are consistent or synchronized; S2.2. Convert all pedestrian motion trajectory data and vehicle location information data into a unified coordinate system, such as from the geographic coordinate system (latitude and longitude) to a plane rectangular coordinate system (such as UTM). Since the motion trajectories of pedestrians and vehicles occur in the same geographic space, these trajectories need to be projected into the same coordinate system for easy comparison and fusion; S2.3. Use the interpolation method to time-align the pedestrian's motion trajectory data and the vehicle's position information data. The data collection frequencies of pedestrians and vehicles may be different, so it is necessary to ensure that the two data streams are aligned in time, that is, for each time point, there is corresponding pedestrian and vehicle position information; S2.4. The coordinates of pedestrians and vehicles are fused by weighted averaging to obtain the fused coordinates: ; ; in, Represents the weight coefficient of pedestrian motion trajectory data, Represents the weight coefficient of vehicle location information data, Represents the horizontal coordinate of the fused data. Represents the vertical coordinate of the fused data. represents the position coordinates of the vehicle, Indicates the time point that needs to be interpolated.

[0020] S3, extract the trajectories of pedestrians and vehicles from the fused data and process them using the dynamic time warping (DTW) algorithm; In this embodiment, the dynamic time warping (DTW) algorithm is a technique for measuring the similarity between two time series. It aligns the two sequences by nonlinearly stretching or compressing the time axis so that they can be compared when they are of different lengths and speeds; pedestrians and vehicles usually have different speeds, even if they move along similar paths. The DTW algorithm can find the optimal time alignment between the two sequences, even if their speeds are inconsistent, allowing effective trajectory matching at different speeds; since pedestrians and vehicles may not start or end their movements at the same time, or may temporarily stop due to reasons such as traffic lights, this will cause a temporal offset in the trajectory data. DTW can eliminate this offset by nonlinearly stretching or compressing the time axis to ensure that the corresponding behavior fragments are compared; the number of trajectory points of pedestrians may not be equal to the number of trajectory points of vehicles, especially in data recorded over a long period of time. DTW can match trajectories of different lengths by finding the best path to ensure correct comparison even if the lengths are different; The trajectories of pedestrians and vehicles are extracted from the fused data and processed using the dynamic time warping (DTW) algorithm, which includes the following steps: S3.1, split the fused continuous data stream into independent trajectory segments according to the time period t1 to be analyzed, ensure that all trajectory data use the same coordinate system, normalize the data to eliminate the influence of scale differences, and preprocess the trajectory segments, including noise reduction and smoothing; S3.2. The distance between trajectory points is calculated by the Euclidean distance method. The Euclidean distance method is optimized for the waiting time of traffic lights on urban roads. The Euclidean distance method is further optimized by considering the impact of traffic flow on the waiting time of traffic lights. Furthermore, the distance between trajectory points is calculated using the Euclidean distance method: ; in, represents the pedestrian trajectory point, represents the vehicle trajectory point, represents the distance between the pedestrian trajectory point and the vehicle trajectory point, Represents the timestamp corresponding to the i-th trajectory point in the pedestrian trajectory data; Represents the vehicle trajectory data The timestamp corresponding to each trajectory point; On urban roads, the control of traffic lights causes vehicles and pedestrians to stop and wait during certain time periods. When vehicles or pedestrians encounter a red light, their movement trajectory will pause instead of continuing to move. If the waiting time of the traffic light is ignored, the dynamic time warping (DTW) algorithm will regard this waiting time as an unnecessary time difference, which will affect the alignment and similarity calculation of the trajectory, resulting in incorrect trajectory matching results; traditional Euclidean distance calculation only considers the difference in spatial coordinates, but ignores the difference in time, especially the waiting or pause between different time points. The existence of traffic lights makes vehicles and pedestrians stationary at specific times, and Euclidean distance calculation will generate unnecessary distance penalties for these stationary states. For example, two trajectory points (one belonging to a pedestrian and the other to a vehicle) may stay in the same position at the same time because of the waiting time of the traffic light. The traditional Euclidean distance calculation will mistakenly believe that there is a large difference between the two trajectory points, resulting in the DTW algorithm failing to match the correct trajectory; The Euclidean distance method is optimized for the waiting time of traffic lights on urban roads: ; in, It represents the distance between the trajectory points of pedestrians and vehicles after considering the waiting time of traffic lights on urban roads; Indicates the degree of influence of pedestrian waiting time on the total distance; Indicates the degree of influence of vehicle waiting time on the total distance; represents the waiting time of pedestrians due to traffic lights; Indicates the waiting time of vehicles due to traffic lights; Traffic light waiting time is a common delay factor in urban traffic, and this delay is usually closely related to traffic flow. During high-flow periods, the waiting time for traffic lights is often longer, while during low-flow periods, the waiting time is shorter. Therefore, the waiting time for traffic lights itself is a factor that has a greater impact on trajectory analysis. Simply using traditional Euclidean distance calculations may ignore this reality. Considering the impact of traffic flow on the waiting time for traffic lights, the actual behavior of pedestrians and vehicles under different traffic conditions can be simulated more accurately, making the reconstruction of trajectories more in line with actual traffic conditions. In real urban traffic, traffic flow will fluctuate greatly in different time periods, which in turn affects the switching time and waiting time of traffic lights. By introducing the influence of traffic flow, the model can dynamically adjust when responding to different traffic conditions to reflect more accurate behavior patterns. For example, during high-flow periods, the relative trajectories of pedestrians and vehicles may shift due to long waits for traffic lights. This change is crucial for identifying traffic anomalies. Considering the impact of traffic flow on the waiting time of traffic lights, the Euclidean distance method is further optimized: ; ; in, represents the distance between the pedestrian trajectory point and the vehicle trajectory point after further optimization; represents a function reflecting the traffic flow situation; represents the weight coefficient of traffic flow; The weight coefficient representing the speed change; represents the weight coefficient of traffic density; Indicates average speed; It indicates a reference speed, such as the average speed during the same period in history or the ideal driving speed for the road section; Indicates the number of vehicles; Indicates the area of ​​the region; S3.3, the cumulative cost matrix is ​​calculated recursively based on the distance between the trajectory points. The main purpose of calculating the cumulative cost matrix by recursion is to find the optimal alignment path between the two time series, so as to minimize the total distance or difference between them; Furthermore, the cumulative cost matrix is ​​calculated recursively as follows: ; in, represents the minimum cumulative cost; Indicates that from the sequence point to another sequence The cumulative cost of each point; Represents the distance from the i-th point of a sequence to the i-th point of another sequence. The cumulative cost of each point; Indicates that from the sequence point to another sequence The cumulative cost of each point; S3.4. Starting from the lower right corner of the cumulative cost matrix, find the optimal alignment path by tracing back the path with the minimum cost. The tracing rule is: Backtrack to the smallest adjacent value: , until you trace back to the upper left corner ; S3.5. Align the corresponding points in the pedestrian trajectory and the vehicle trajectory by tracing the optimal alignment path, and obtain the reconstructed trajectory of the pedestrian trajectory and the vehicle trajectory in time and space.

[0021] S4, based on the support vector machine model, the spatiotemporal patterns of pedestrian and vehicle trajectories are modeled to identify road traffic anomalies, and for traffic congestion, a traffic congestion function is introduced in the modeling process; In this embodiment, the support vector machine (SVM) is a supervised learning model used for classification and regression analysis. It achieves efficient data classification by finding an optimal hyperplane to maximize the interval between data points of different categories. By analyzing the spatiotemporal behavior patterns of pedestrians and vehicles, potential safety hazards can be discovered in a timely manner, such as illegal red light running, illegal lane changes, or abnormal behaviors such as speeding. This helps prevent traffic accidents and protect the lives and property of road users. The SVM model can automatically detect violations of traffic rules and provide law enforcement officers with reliable evidence support. This not only improves law enforcement efficiency, but also has a deterrent effect, prompting drivers and pedestrians to comply with traffic regulations. Based on the support vector machine model, the spatiotemporal patterns of pedestrian and vehicle trajectories are modeled to identify road traffic anomalies, including the following steps: S4.1. Extract and analyze the interaction features between pedestrians and vehicles. The interaction features include: interaction distance features (calculating the relative distance between pedestrians and vehicles at the same time point), speed and acceleration matching features (calculating the speed difference between pedestrians and vehicles at the interaction point), extract spatiotemporal features from the fused pedestrian and vehicle trajectory data, and the spatiotemporal features include trajectory speed features, trajectory acceleration features, trajectory curvature features, and trajectory timing features, and preprocess the extracted spatiotemporal features and interaction features; S4.2. Based on the preprocessed spatiotemporal features and interaction features, select features related to identifying abnormal behaviors, including trajectory velocity features and trajectory acceleration features, and convert the data into a format acceptable to the support vector machine model; S4.3, taking the spatiotemporal features and interaction features of each trajectory segment as sample input, constructing a training dataset and a test dataset; S4.4. Select a kernel function (Radial Basis Kernel (RBF)) to fit the data, optimize the kernel function for traffic congestion, and use the support vector machine model to perform classification training for abnormal behavior. The training goal is to build a classifier that can predict whether the behavior is normal or abnormal based on the input trajectory features; Furthermore, the kernel function (Radial Basis Kernel (RBF)) measures the similarity between two data points in high-dimensional space by calculating the exponential decay of the square of the distance between them. The main purpose of choosing the Radial Basis Function (RBF) kernel for data fitting is to improve the performance and flexibility of the Support Vector Machine (SVM) model when dealing with nonlinear classification problems. The RBF kernel can map the data in the original input space to a high-dimensional feature space, in which even data that is originally linearly inseparable in the low-dimensional space may become linearly separable. This is especially important for complex spatiotemporal patterns such as pedestrian and vehicle trajectories, because these patterns are usually not simple linear relationships; the RBF kernel introduces nonlinear transformations, allowing the SVM model to capture the complex relationships between data. This helps to improve the model's ability to distinguish different traffic behavior patterns, especially when the boundary between abnormal behavior and normal behavior is not obvious or very complex; The kernel function is: ; in, Represents a given input point and The similarity calculated between Represents the kernel width, which controls the smoothness of the kernel function; The feature vector representing the pedestrian trajectory data point; The feature vector representing the vehicle trajectory data point; When analyzing the trajectories of pedestrians and vehicles, traffic jams can cause significant changes in the way pedestrians and vehicles move. Pedestrians and vehicles may be stagnant for a long time on a congested road, which does not mean that they are no longer similar. Standard kernel functions (such as Gaussian kernels) only consider spatial distances and ignore stagnation in time or changes in traffic flow. In the case of traffic jams, ignoring these factors will lead to inaccurate trajectory matching and incorrect similarity calculations. For example, two points (one is a pedestrian and the other is a vehicle) may stay in the same position due to traffic lights or traffic jams. Traditional kernel functions may mistakenly believe that there is a big difference between the two points, resulting in incorrect trajectory alignment; by modeling and optimizing traffic jams, the spatiotemporal patterns of traffic flow can be extracted more accurately. Models such as support vector machines (SVMs) rely on similarity calculations of input data for pattern recognition. Accurate trajectory alignment is crucial, especially when identifying abnormal behaviors (such as traffic accidents). Optimizing kernel functions can improve the accuracy of spatiotemporal pattern modeling and help identify abnormal behaviors caused by traffic jams (such as traffic accidents, congestion, etc.), thereby achieving more effective urban governance; Optimize the kernel function for traffic congestion: ; In the governance scenario, the physical distance between two vehicles is indeed important, but factors such as vehicle speed, traffic signals, and traffic flow also affect the "similarity" between the two vehicles. For example, if two vehicles are both stopped at a traffic light, although their distances are similar, their behaviors are not exactly the same, so it is necessary to consider the differences in the behavioral level, and the interactive feature items Defined as a set of additional factors that affect similarity, which not only depends on physical distance, but also includes other influencing factors (speed difference, signal light status, traffic flow). By introducing these factors, the model can more accurately capture the interactive influence relationship between people and vehicles (at traffic intersections, schools, and near shopping districts, pedestrian flow is high, and vehicles may need to frequently avoid or slow down; on highways or dedicated lanes, vehicle flow is high, while pedestrian flow is usually zero). For example, the difference in traffic flow may affect the behavior pattern between two vehicles. Even if the physical distance between them is similar, if the traffic difference is large, their driving pattern or behavior may be very different, so their similarity should be different; ; ; in, represents the similarity calculated after optimization for traffic congestion; represents the weight of controlling the impact of traffic congestion on similarity; represents the traffic congestion function, i.e., the distance of the interaction feature; represents the weight coefficient for controlling speed difference; Indicates the weight coefficient for controlling the state of the traffic light; Represents the weight coefficient for controlling traffic flow; Indicates the speed difference between pedestrians and vehicles; Represents the influence function of traffic lights on similarity; Indicates traffic flow differences; S4.5. Use the trained support vector machine model to determine whether the new trajectory data is abnormal. By combining the trajectory characteristics of the two, it can be determined whether there are abnormal interactions or behaviors in the environment where people and vehicles coexist, such as vehicles not driving according to rules or pedestrians taking the wrong road. Furthermore, the support vector machine model is: ; in, Represents the decision function output of the support vector machine model, which is used to Make a prediction to decide which class a data point belongs to; represents the number of support vectors; represents the Lagrange multiplier; Indicates The category labels of training samples; Represents the index of the training sample; Represents the bias term.

[0022] Example 2: This example provides an urban governance data fusion analysis system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement any one of the urban governance data fusion analysis methods described above.

[0023] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited by the above embodiments, and the above embodiments and descriptions are only preferred examples of the present invention, and are not intended to limit the present invention. Without departing from the spirit and scope of the present invention, the present invention may have various changes and improvements, and these changes and improvements all fall within the scope of the present invention to be protected.

Claims

1. A method for fusion analysis of urban governance data, characterized in that: The following steps are involved: S1. Use sensors to obtain the movement trajectory data of pedestrians in the city, and use GPS devices to obtain the location information data of vehicles in the city; S2, merging the pedestrian's motion trajectory data and the vehicle's location information data through a weighted average method; S3, extract the trajectories of pedestrians and vehicles from the fused data and process them using the dynamic time warping algorithm; S4. Based on the support vector machine model, the spatiotemporal patterns of pedestrian and vehicle trajectories are modeled to identify road traffic anomalies, and for traffic congestion, a traffic congestion function is introduced in the modeling process.

2. The urban governance data fusion analysis method according to claim 1 is characterized by: In S1, the pedestrian's motion trajectory data includes: position coordinates , Timestamp .

3. The urban governance data fusion analysis method according to claim 2 is characterized by: In S2, the pedestrian's motion trajectory data and the vehicle's location information data are fused by weighted averaging method, including the following steps: S2.1, preprocessing the motion trajectory data of pedestrians and the location information data of vehicles; S2.2, converting the motion trajectory data of all pedestrians and the position information data of vehicles into a unified coordinate system; S2.3, using interpolation method to time align the motion trajectory data of the pedestrian and the position information data of the vehicle; S2.

4. The coordinates of pedestrians and vehicles are fused by weighted averaging to obtain the fused coordinates: ； ； in, Represents the weight coefficient of pedestrian motion trajectory data, Represents the weight coefficient of vehicle location information data, Represents the horizontal coordinate of the fused data. Represents the vertical coordinate of the fused data. represents the position coordinates of the vehicle, Indicates the time point that needs to be interpolated.

4. The urban governance data fusion analysis method according to claim 3 is characterized by: In S3, the trajectories of pedestrians and vehicles are extracted from the fused data and processed using a dynamic time warping algorithm, including the following steps: S3.1, dividing the fused continuous data stream into independent trajectory segments according to the time period t1 to be analyzed, and preprocessing the trajectory segments; S3.

2. The distance between trajectory points is calculated by the Euclidean distance method. The Euclidean distance method is optimized for the waiting time of traffic lights on urban roads. The Euclidean distance method is further optimized by considering the impact of traffic flow on the waiting time of traffic lights. S3.3, calculating the cumulative cost matrix by recursion based on the distance between trajectory points; S3.

4. Starting from the lower right corner of the cumulative cost matrix, find the optimal alignment path by tracing back the path with the minimum cost; S3.

5. Align the corresponding points in the pedestrian trajectory and the vehicle trajectory by tracing the optimal alignment path, and obtain the reconstructed trajectory of the pedestrian trajectory and the vehicle trajectory in time and space.

5. The urban governance data fusion analysis method according to claim 4 is characterized by: In S3.3, the distance between trajectory points is calculated using the Euclidean distance method: ； in, represents the pedestrian trajectory point, represents the vehicle trajectory point, represents the distance between the pedestrian trajectory point and the vehicle trajectory point, represents the first The timestamp corresponding to each trajectory point; Represents the vehicle trajectory data The timestamp corresponding to each trajectory point; The Euclidean distance method is optimized for the waiting time of traffic lights on urban roads: ； in, It represents the distance between the trajectory points of pedestrians and vehicles after considering the waiting time of traffic lights on urban roads; Indicates the degree of influence of pedestrian waiting time on the total distance; Indicates the degree of influence of vehicle waiting time on the total distance; represents the waiting time of pedestrians due to traffic lights; Indicates the waiting time of vehicles due to traffic lights; Considering the impact of traffic flow on the waiting time of traffic lights, the Euclidean distance method is further optimized: ； in, represents the distance between the pedestrian trajectory point and the vehicle trajectory point after further optimization; represents a function reflecting the traffic flow situation; Represents the weight coefficient of traffic flow.

6. The urban governance data fusion analysis method according to claim 5 is characterized by: In S3.3, the cumulative cost matrix is ​​calculated by recursion as follows: ； in, represents the minimum cumulative cost; Indicates that from the sequence point to another sequence The cumulative cost of each point; Indicates that from the sequence point to another sequence The cumulative cost of each point; Indicates that from the sequence point to another sequence The cumulative cost of each point.

7. The urban governance data fusion analysis method according to claim 6 is characterized by: In S4, identifying road traffic anomalies by modeling the spatiotemporal patterns of pedestrian and vehicle trajectories based on a support vector machine model includes the following steps: S4.1, extract and analyze the interaction features between pedestrians and vehicles, extract spatiotemporal features from the fused pedestrian and vehicle trajectory data, and preprocess the extracted spatiotemporal features and interaction features; S4.2, based on the preprocessed spatiotemporal features and interaction features, select features related to identifying abnormal behaviors; S4.3, taking the spatiotemporal features and interaction features of each trajectory segment as sample input, constructing a training dataset and a test dataset; S4.4, select the kernel function to fit the data, optimize the kernel function for traffic congestion, and use the support vector machine model to classify abnormal behaviors; S4.

5. Use the trained support vector machine model to determine whether the new trajectory data is abnormal.

8. The urban governance data fusion analysis method according to claim 7 is characterized by: In S4.4, the kernel function is: ； in, Represents a given input point and The similarity calculated between Represents the kernel width, which controls the smoothness of the kernel function; The feature vector representing the pedestrian trajectory data point; The feature vector representing the vehicle trajectory data point; Optimize the kernel function for traffic congestion: ； ； in, represents the similarity calculated after optimization for traffic congestion; represents the weight of controlling the impact of traffic congestion on similarity; represents the traffic congestion function; represents the weight coefficient for controlling speed difference; Indicates the weight coefficient for controlling the state of the traffic light; Represents the weight coefficient for controlling traffic flow; Indicates the speed difference between pedestrians and vehicles; Represents the influence function of traffic lights on similarity; Represents the difference in traffic volume.

9. The urban governance data fusion analysis method according to claim 8 is characterized by: In S4.5, the support vector machine model is: ； in, Represents the decision function output of the support vector machine model; represents the number of support vectors; represents the Lagrange multiplier; Indicates The category labels of the training samples; Represents the index of the training sample; Represents the bias term.

10. An urban governance data fusion analysis system, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: The processor executes a computer program to implement the urban governance data fusion and analysis method as described in any one of claims 1-9.

Images