Traffic comprehensive law enforcement situation awareness, study and judgment system based on big data and cloud computing

Abstract

The invention discloses a traffic comprehensive law enforcement situation awareness, study and judgment system based on big data and cloud computing, and relates to the technical field of traffic law enforcement management. The problem that a traffic law enforcement platform based on edge calculation has calculation and storage bottlenecks and cannot timely and accurately recognize novel and rare traffic illegal behaviors when dealing with extremely complex traffic scenes and mass data is solved. According to the invention, the law enforcement scene and the vehicle abnormal condition are identified through data collected by various devices, illegal behaviors are identified and early warned in time by combining audio and video analysis, the law enforcement accuracy and efficiency are improved, the road safety and the market order are guaranteed, the traffic law enforcement situation model carries out clustering analysis on traffic data, a visual situation map is generated, and law enforcement decision is assisted. Resource configuration is optimized, illegal trend is predicted, a targeted scheme is formulated, law enforcement scientificity is improved, the system is optimized by analyzing feedback information of law enforcement officers, key information is pushed in real time, collaborative law enforcement is promoted, and comprehensive law enforcement efficiency is improved.

Description

Technical Field

[0001] This invention relates to the field of traffic law enforcement management technology, and in particular to a traffic integrated law enforcement situation awareness and analysis system based on big data and cloud computing. Background Technology

[0002] Traditional law enforcement methods have many limitations, relying on manual patrols and simple equipment monitoring, making it difficult to achieve real-time and comprehensive monitoring of traffic violations across all scenarios. Faced with increasing traffic flow and complex and diverse violations, Chinese patent application CN119274349 discloses a digital traffic enforcement insight engine and analysis platform based on edge computing. Through information association and verification, it ensures the accuracy and integrity of data, providing reliable standard traffic analysis data for comprehensive transportation administrative law enforcement agencies, supporting more efficient decision-making and execution. The platform can automatically extract, fuse, associate, and verify data, reducing manual intervention, improving management efficiency, and reducing the risk of human error. Through automated analysis processes and intelligent analysis, the platform reduces the burden of manual intervention and analysis, improving the efficiency and accuracy of decision support. Accelerated computing via chips further improves processing efficiency and speed, enabling the system to handle the real-time processing needs of large-scale data streams and allowing for adjustments to data processing flows and analysis models based on actual requirements.

[0003] While the aforementioned patents address the problems of traditional law enforcement methods, the computing and storage capabilities of edge devices may reach a bottleneck when dealing with extremely complex traffic scenarios and massive amounts of data, leading to a decline in system performance. This may prevent timely and accurate identification and assessment of some new and rare traffic violations, thus limiting the effectiveness of the patents in the overall traffic law enforcement system. Summary of the Invention

[0004] The purpose of this invention is to provide a traffic comprehensive law enforcement situation awareness and analysis system based on big data and cloud computing. By accurately identifying law enforcement scenarios and vehicle anomalies, it can promptly warn of illegal and irregular behaviors, and use a traffic law enforcement situation model to deeply analyze traffic data, thereby achieving more targeted law enforcement and effectively safeguarding road safety and market order, thus solving the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A comprehensive traffic law enforcement situation awareness and analysis system based on big data and cloud computing includes: The law enforcement scene monitoring module is configured to acquire traffic data in law enforcement scenes in real time, extract vehicle feature data, record audio and video data, and perform law enforcement scene recognition, determine feature parameters, and judge whether the vehicles in the law enforcement scene are abnormal. The situational awareness analysis module is configured to analyze the traffic law enforcement situation based on the output of the law enforcement scenario monitoring module and using a traffic law enforcement situational model. The law enforcement decision support module is configured to optimize the allocation of law enforcement resources based on the results of traffic law enforcement situation analysis, predict violation trends and key areas based on historical and real-time data, and formulate law enforcement plans. The law enforcement feedback interaction module is configured to receive feedback information from law enforcement personnel, classify and analyze it, and push law enforcement scenario information, early warning information and law enforcement plans to law enforcement personnel in real time.

[0006] Furthermore, the law enforcement scene monitoring module performs law enforcement scene identification, specifically including: Extract video stream data from traffic data in law enforcement scenarios and preprocess it to obtain preprocessed video stream data; Feature extraction is performed on the preprocessed video stream data, and an initial feature set for traffic data is obtained based on the extraction results; Extract key features relevant to the law enforcement scenario from the initial feature set and integrate them into a subset of key features; Obtain information on the topology of the traffic network, pre-defined law enforcement management methods, and the node attributes of each traffic node; Based on the node attributes and topology information of each traffic node, as well as the preset law enforcement management method information, determine the traffic topology weight value of each traffic node; Based on the traffic topology weight value of each traffic node, the basic values ​​of the relevant indicators for each law enforcement scenario in the traffic data are calculated. Obtain the temporal feature information corresponding to traffic data in law enforcement scenarios, and extract the temporal sequence data of relevant indicators for each law enforcement scenario from the traffic data based on the temporal feature information; The law enforcement scenario characteristic information of each indicator is determined based on the time series data of the relevant indicators for each law enforcement scenario; The time-series data and basic values ​​of each law enforcement scenario-related indicator are used as input samples for the model, while the law enforcement scenario feature information of each indicator is used as output samples for the model. The preset network model is trained to obtain the recognition model of each law enforcement scenario-related indicator. The identification model, based on the relevant indicators of each law enforcement scenario, identifies target features in the video stream data, determines the current law enforcement scenario type, and sets corresponding law enforcement scenario labels.

[0007] Furthermore, based on the node attributes and topology information of each traffic node, as well as the preset law enforcement management method information, the traffic topology weight value of each traffic node is determined, including: Extract the traffic flow corresponding to the key time nodes contained in the node attributes of each traffic node; The average traffic flow is obtained based on the traffic flow corresponding to the key time points contained in the node attributes of each traffic node; Extract the topology information of each traffic node, wherein the topology information includes the number of traffic nodes with which each traffic node has a corresponding relationship and the percentage change in traffic volume of the traffic nodes with which each traffic node has a corresponding relationship when the traffic volume of each traffic node increases by a percentage. The average percentage change in traffic flow of each traffic node is obtained by using the percentage change in traffic flow of its associated traffic nodes corresponding to the percentage increase in traffic flow of each traffic node. The average traffic flow at each traffic node is compared with a preset traffic flow reference value to obtain the ratio between the average traffic flow at each traffic node and the preset traffic flow reference value. The ratio between the average traffic flow of each traffic node and the preset traffic flow reference value is compared with the average percentage change of the associated traffic nodes corresponding to each traffic node; The traffic topology weight value of each traffic node is set based on the comparison between the ratio of the average traffic flow of each traffic node to the preset traffic flow reference value and the average percentage change of the associated traffic nodes corresponding to each traffic node, combined with the preset law enforcement management method information corresponding to each traffic node.

[0008] Furthermore, based on the comparison between the ratio of the average traffic flow of each traffic node to a preset traffic flow reference value and the average percentage change of associated traffic nodes corresponding to each traffic node, combined with the preset law enforcement management method information corresponding to each traffic node, the traffic topology weight value of each traffic node is set, including: When the ratio between the average traffic flow of each traffic node and the preset traffic flow reference value exceeds the average percentage change of the associated traffic nodes corresponding to each traffic node, the traffic node whose ratio between the average traffic flow of each traffic node and the preset traffic flow reference value exceeds the average percentage change of the associated traffic nodes corresponding to each traffic node is designated as the first target traffic node. Retrieve the traffic capacity and the number of branch roads corresponding to each first target traffic node; Extract the average traffic flow for each branch road contained in each first target traffic node; The average traffic flow of each branch road included in each first target traffic node is compared with the traffic capacity of the first target traffic node to obtain the traffic flow percentage value of each branch road included in each first target traffic node. The traffic flow percentage of each branch road is compared with the average percentage change of the associated traffic nodes corresponding to each first target traffic node. The branch roads whose traffic flow percentage is not lower than the average percentage change of the associated traffic nodes are extracted as target branch roads. The traffic topology weight value corresponding to the first target traffic node is obtained by combining the traffic flow percentage of the target intersection road with the preset topology weight benchmark value. The traffic topology weight value corresponding to the first target traffic node is obtained by the following formula:

[0009] Among them, Y 01 Y0 represents the traffic topology weight value corresponding to the first target traffic node; Y0 represents the preset topology weight baseline value; B 01 This represents the average percentage change of associated traffic nodes corresponding to the first target traffic node; n represents the number of target branch roads corresponding to the first target traffic node; L 01i This represents the percentage of traffic flow corresponding to the i-th target intersection road at the first target traffic node.

[0010] Furthermore, based on the comparison between the ratio of the average traffic flow of each traffic node to a preset traffic flow reference value and the average percentage change of associated traffic nodes corresponding to each traffic node, combined with the preset law enforcement management method information corresponding to each traffic node, the traffic topology weight value of each traffic node is set, which also includes: When the ratio between the average traffic flow of each traffic node and the preset traffic flow reference value does not exceed the average percentage change of the associated traffic nodes corresponding to each traffic node, the traffic node whose ratio between the average traffic flow of each traffic node and the preset traffic flow reference value does not exceed the average percentage change of the associated traffic nodes corresponding to each traffic node is taken as the second target traffic node. Extract the preset law enforcement management method information for each second target traffic node, wherein the law enforcement management method information includes the law enforcement resource distribution intensity coefficient corresponding to each fork road included in the second target traffic node, wherein the law enforcement resource distribution intensity is divided into strong, medium and weak; and the law enforcement resource distribution intensity coefficients corresponding to the strong, medium and weak levels of the law enforcement resource distribution intensity are 1.0, 0.8 and 0.5, respectively. Retrieve the traffic capacity and the number of branch roads corresponding to each second target traffic node; Extract the average traffic flow for each branch road contained in each second target traffic node; The traffic flow ratio of each branch road included in each second target traffic node is obtained by comparing the average traffic flow of each branch road with the traffic capacity of the second target traffic node. The traffic flow percentage of each branch road is compared with the average percentage change of the associated traffic nodes corresponding to each second target traffic node. The branch roads whose traffic flow percentage is not lower than the average percentage change of the associated traffic nodes are extracted as target branch roads. The traffic topology weight value corresponding to the second target traffic node is obtained by combining the traffic flow ratio corresponding to the target intersection road with the law enforcement resource distribution intensity coefficient corresponding to the target intersection road and the preset topology weight benchmark value. The traffic topology weight value corresponding to the second target traffic node is obtained by the following formula:

[0011] Among them, Y 02 Y0 represents the traffic topology weight value corresponding to the second target traffic node; B represents the preset topology weight baseline value. 02 This represents the average percentage change of associated traffic nodes corresponding to the second target traffic node; m represents the number of target branch roads corresponding to the second target traffic node; L 02i This represents the traffic flow percentage corresponding to the i-th target intersection road at the second target traffic node; k i This represents the intensity coefficient of law enforcement resource distribution corresponding to the i-th target intersection road corresponding to the second target traffic node.

[0012] Furthermore, the law enforcement scene monitoring module, in its law enforcement scene identification process, also includes: By using the identification model of relevant indicators for each law enforcement scenario, the target law enforcement scenario features corresponding to the target time series data of each key feature in the key feature subset are obtained; Based on the target law enforcement scenario features of each key feature, obtain the first law enforcement scenario feature for each law enforcement scenario; Obtain the changes in the target law enforcement scenario characteristics of each law enforcement scenario in the traffic data of the law enforcement scenario, and determine the law enforcement scenario change rules for each law enforcement scenario based on the changes. Enforcement scenarios with a similarity to the rule of change of enforcement scenarios greater than or equal to a preset threshold are identified as the same type of enforcement scenario. The second target enforcement scenario feature of any enforcement scenario in each type of enforcement scenario is identified as the final enforcement scenario feature of that type of enforcement scenario. Furthermore, the feature parameters are determined based on the law enforcement scenario identification results, which include: overloaded / oversized scenarios, illegal operation scenarios, and "two-passenger-one-dangerous-one-heavy" scenarios. Based on the overloaded / oversized scenarios, for truck-type vehicles, feature parameters such as axle group type, number of axles, weighing data, and length, width, and height data are determined. Based on the illegal operation scenarios, feature parameters such as operating license, business scope, and license validity period are determined. Based on the "two-passenger-one-dangerous-one-heavy" scenarios, parameter data such as vehicle operation qualification compliance mark and the set of coordinates of the prescribed driving route are determined.

[0013] Furthermore, determining whether there are any anomalies in vehicles within the law enforcement scenario specifically includes: When the axle weight of a vehicle exceeds the vehicle axle weight threshold in the real-time traffic data of law enforcement scenarios, and the change in vehicle height exceeds the range of vehicle height change, combined with abnormal sounds in audio and video data, it is determined that there may be overloading behavior. When the frequency of passenger boarding and alighting exceeds the passenger boarding and alighting frequency threshold and the operating time exceeds the operating time range, the voice data of audio and video data is analyzed to determine whether the conversation between the passenger and the driver involves operation-related information. If there is evidence to support illegal operation, it is judged as suspected illegal operation. If the vehicle operation qualification mark does not conform to the vehicle operation qualification compliance mark, or the driving route coordinates are not within the range of the prescribed driving route coordinate set, then it is determined that the "two passenger vehicles, one dangerous goods vehicle and one heavy vehicle" are in violation. Once an anomaly is detected in a vehicle within a law enforcement scenario, an early warning mechanism is immediately triggered, generating warning information based on the type of law enforcement scenario and the category of anomaly, combined with audio and video data from the law enforcement process.

[0014] Furthermore, the traffic enforcement situation model specifically includes: Monitor the temporal and spatial distribution of traffic data in law enforcement scenarios and obtain detection results; Based on the detection results, cluster analysis is performed on data from different law enforcement scenarios to form different situation categories, determine the clustering characteristics of various law enforcement scenarios, and the changing direction and degree trend of each situation category; The correlation between traffic violations and traffic flow, weather conditions, time, and accident information is determined, and a visualized traffic enforcement situation map is generated based on the correlation, thereby constructing a traffic enforcement situation model. The traffic enforcement situation map visually marks different road sections according to their traffic enforcement situation categories, and uses icons of different sizes to represent the traffic flow. At the same time, weather information, time information, and accident location and type are overlaid on the traffic enforcement situation map.

[0015] Furthermore, the classification, organization, and analysis of feedback information from law enforcement personnel also include: Text information analysis is performed on the feedback information from law enforcement personnel, and the text information is segmented into multi-level topic trees; The text information is segmented into words, and similarity matching is performed based on the constructed thesaurus of synonyms in the field of traffic law enforcement and each extracted word; Based on the similarity matching results, duplicates are removed to obtain the target word set, and keywords are extracted from the target word set. Based on the keyword extraction results from the relevant feedback information clusters of each law enforcement scenario, determine the keywords included in each layer of the topic tree; Different categories are distinguished: feedback on the accuracy of law enforcement scenario judgments, feedback on the effectiveness of law enforcement decision implementation, and suggestions for system function improvement.

[0016] Compared with the prior art, the beneficial effects of the present invention are: By collecting data from multiple devices, the system accurately identifies law enforcement scenarios and vehicle anomalies. It extracts multi-dimensional features such as vehicle shape, movement status, and abnormal sounds from video stream data, constructs an initial feature set, and filters key features, significantly improving the ability to identify illegal and irregular behaviors. In illegal operation scenarios, it combines passenger boarding and alighting frequency, operation duration, and human voice data in audio and video to judge operation behavior, with an accuracy far exceeding that of traditional methods.

[0017] By clustering and analyzing data from different enforcement scenarios using a traffic enforcement situation model, the correlation between traffic violations and traffic flow, weather, time, and accident information is uncovered. This provides a comprehensive basis for enforcement decisions, enabling timely responses to complex situations and accurate identification of new and rare traffic violations. It overcomes the limitations of traditional edge computing platforms. In enforcement scenario identification, traffic topology weight values ​​are determined by comprehensively considering traffic node attributes, topology structure, and preset enforcement management methods, achieving accurate assessment of the importance of traffic nodes. Weights are determined in conjunction with the distribution of enforcement resources, providing a scientific basis for the rational allocation of enforcement resources, optimizing enforcement decisions, and improving the efficiency of enforcement resource utilization. This is a refined management method that traditional enforcement methods lack. Attached Figure Description

[0018] Figure 1 This is a module diagram of the traffic integrated law enforcement situation awareness and analysis system based on big data and cloud computing of the present invention. Detailed Implementation

[0019] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] To address the technical challenges faced by edge computing-based traffic enforcement platforms in handling extremely complex traffic scenarios and massive amounts of data, including computational and storage bottlenecks, the inability to promptly and accurately identify new and rare traffic violations, and the difficulty in achieving real-time and comprehensive monitoring of traffic violations across all scenarios, please refer to [link to relevant documentation]. Figure 1 This embodiment provides the following technical solution: A comprehensive traffic law enforcement situation awareness and analysis system based on big data and cloud computing includes: The law enforcement scenario monitoring module is used for: Based on monitoring equipment, including traffic cameras, geomagnetic sensors, radar, and loop detectors, real-time traffic data in law enforcement scenarios is acquired, such as traffic flow (vehicle flow, pedestrian flow, vehicle type composition, etc.), speed (average speed of road segment, instantaneous speed, vehicle speed distribution, etc.), and accident information (accident type, accident location, accident time, casualties, etc.). At the same time, vehicle feature data, including the real-time location and movement status of vehicles, is extracted from the traffic data in law enforcement scenarios. This is obtained through mobile data sources such as vehicle-mounted and mobile phone signaling, and audio and video data during the law enforcement process is recorded. Enforcement scenario identification is performed on traffic data in enforcement scenarios. Based on the enforcement scenario identification results, corresponding feature parameters are determined, and it is determined whether there are any anomalies in the vehicles within the enforcement scenario. The enforcement scenario identification results include: overloaded / oversized scenarios, illegal operation scenarios, and "two types of passenger vehicles, one type of dangerous goods vehicle, and one type of heavy vehicle" scenarios. Based on the overloaded / oversized scenario, for truck-type vehicles, characteristic parameters such as axle group type, number of axles, weighing data, and length, width, and height data are determined to determine whether the load exceeds the prescribed limit or the dimensions are excessive. Based on the illegal operation scenario, characteristic parameters such as the operating license, business scope, and validity period of the license are determined. Based on the "two types of passenger vehicles, one type of dangerous goods vehicle, and one type of heavy vehicle" scenario, parameter data such as vehicle operation qualification compliance identification and the set of coordinates for the prescribed driving route are determined. The situational awareness analysis module is used to analyze the traffic enforcement situation based on the enforcement scene identification results, anomaly judgment results, and traffic data of the enforcement scene output by the enforcement scene monitoring module, using a traffic enforcement situational model. In this embodiment, spatial clustering algorithms are used to perform clustering analysis on traffic flow, frequency of violations, and other data under different law enforcement scenarios. Based on the density distribution of data points, different clusters are automatically identified. For example, in the traffic data of a certain city, the spatial clustering algorithm can cluster road sections with frequent violations during weekday morning and evening rush hours as high-risk law enforcement areas, and road sections with low traffic flow and few violations at night as low-risk areas. This determines the clustering characteristics of various law enforcement scenarios, such as peak traffic flow times and concentration of violation types in high-risk areas, as well as the changing direction and degree of each situation category, such as the fluctuation trend of violation frequency in high-risk areas with seasonal and weekday changes. In this embodiment, the traffic enforcement situation model specifically includes: ①: Screening model for clues of illegal ride-hailing vehicles within 30 days: Using cameras to identify license plates, the data is pushed to the message middleware. The analysis system queries the basic information of the vehicle. If there is no information on the vehicle under the transportation administration and there are ride-hailing platform operation orders within the past 30 days, it is identified as a suspected illegal ride-hailing vehicle and an early warning is sent.

[0021] ②: Suspected illegal passenger transport model: By screening vehicle certificate information through data from the Ministry of Transport and the provincial system's transport administration module, a multi-dimensional analysis is conducted on the highway toll data of non-commercial vehicles that have not obtained a "Road Transport Permit". Based on factors such as more than 30 highway trips per month, cumulative toll mileage exceeding 4,000 kilometers, suspected pick-up and drop-off of passengers along the route less than 15 times, weekday travel frequency between 3 and 8 times, and previous penalties or complaints, an illegal passenger transport characteristic model is constructed. The list of suspected vehicles is dynamically updated, and the vehicles are included in the high-risk vehicle database to be verified, and early warnings are pushed out.

[0022] ③: Screening model for vehicles that are not online within 24 hours (passenger, hazardous materials, and heavy-duty vehicles): Utilizing field sensing equipment such as checkpoints to identify license plates, and comparing vehicle passage records, basic vehicle information, and location information, warnings are sent to operating vehicles with passage records but no location information in the past 24 hours. Simultaneously, warnings from 24 hours prior are re-verified daily, and warnings for vehicles whose location information has been successfully re-transmitted are removed.

[0023] ④: Screening model for overdue annual inspection of commercial vehicles: The field sensing equipment identifies the license plate and pushes the data. The analysis system queries the basic information of the vehicle. If the valid annual inspection period has expired (except for rental vehicles), an early warning clue is pushed.

[0024] ⑤: Screening model for clues of vehicles that have been suspended from operation but are engaged in commercial activities: By recognizing license plates through various field sensing devices (such as checkpoints, non-stop weighing systems, smart cameras, etc.), and comparing vehicle passage records with basic vehicle information, early warning clues are generated for vehicles that are suspended from operation but are engaged in commercial activities.

[0025] ⑥: Dynamic monitoring anomaly screening model for out-of-province hazardous materials transport vehicles: Analyze out-of-province hazardous materials transport vehicles passing through highway toll stations or toll gantries. If the last update time of their satellite positioning data is more than 10 minutes later than the passage time, an early warning record will be generated.

[0026] ⑦: Out-of-province hazardous materials transport vehicles operating in different locations for a long time: Analyze out-of-province hazardous materials transport vehicles that have been operating within the province for more than 90 consecutive days based on satellite positioning data, and generate tracking and early warning records.

[0027] ⑧: Expired road transport permits for out-of-province hazardous materials transport vehicles: Analyze out-of-province hazardous materials transport vehicles that have been operating within the province for more than 90 consecutive days based on satellite positioning data, and generate tracking and early warning records.

[0028] ⑨: Expired annual inspection of dangerous goods transport vehicles from other provinces: Analyze dangerous goods transport vehicles from other provinces that are operating within the province and whose road transport permits have expired, generate early warning records, and dynamically update the list of problematic vehicles.

[0029] The law enforcement decision support module is used to optimize the allocation of law enforcement resources based on the results of traffic law enforcement situation analysis and the characteristics and trends of different law enforcement scenarios. At the same time, based on historical law enforcement data and real-time traffic situation analysis results, it predicts future violation trends and key areas, and formulates targeted law enforcement plans. For example, the prediction model based on time series analysis captures the trend, seasonality and periodicity of time series data. For example, by analyzing the number of violations per week on a certain road segment over the past year, the prediction model can predict the violation trend in the next few weeks, and predict the growth or decline trend of violation events in advance, providing a reference for the early deployment of law enforcement resources. By combining traffic flow data, violation location data and geographic information, it calculates the statistics of each location and identifies the spatial clustering patterns of data. High-value areas are key areas where violations occur frequently. For example, on a city map, the hotspot analysis algorithm can identify intersections and road segments with frequent traffic accidents, as well as areas where illegal operating vehicles are concentrated, providing precise directions for law enforcement personnel to focus on monitoring and allocate law enforcement resources. In this embodiment, based on the high-incidence areas and time periods of violations, and combined with the distribution of law enforcement personnel and equipment, the optimal law enforcement personnel scheduling scheme and law enforcement equipment deployment strategy are calculated using algorithms such as linear programming to improve law enforcement efficiency. In conjunction with traffic regulations and policies, specific law enforcement action suggestions are provided to law enforcement personnel, such as recommending appropriate law enforcement procedures and penalties for different types of violations, to assist law enforcement personnel in making more accurate and efficient law enforcement decisions. The law enforcement feedback interaction module is used to receive feedback information from law enforcement officers during actual law enforcement processes, including feedback on the accuracy of law enforcement scenario judgments, feedback on the effectiveness of law enforcement decision implementation, and suggestions for improving system functions. This feedback is then categorized, organized, and analyzed to serve as the basis for optimizing law enforcement scenario monitoring models, situational awareness analysis algorithms, and law enforcement decision support strategies. Simultaneously, it pushes relevant law enforcement scenario information, early warning information, and law enforcement plans to law enforcement officers in real time, ensuring that they can obtain key information promptly, improving law enforcement response speed, achieving rapid information transmission and collaborative law enforcement, and enhancing the overall efficiency of comprehensive traffic law enforcement.

[0030] In this embodiment, the law enforcement scenario monitoring module monitors various traffic law enforcement scenarios in real time, such as overloading, passenger vehicles, dangerous goods vehicles, and heavy vehicles, as well as illegal operations. It accurately identifies various illegal and irregular behaviors and judges whether vehicles are abnormal based on corresponding feature parameters, providing accurate basis for subsequent law enforcement and effectively improving the accuracy of law enforcement. The situational awareness analysis module uses the traffic law enforcement situational model to conduct in-depth analysis of the law enforcement scenario identification results, anomaly judgment results, and traffic data, and uncovers the potential patterns and trends of traffic law enforcement situation, helping law enforcement departments to fully understand the traffic law enforcement status.

[0031] Specifically, based on the node attributes and topology information of each traffic node, as well as the preset law enforcement management method information, the traffic topology weight value of each traffic node is determined, including: Extract the traffic flow corresponding to the key time nodes contained in the node attributes of each traffic node; The average traffic flow is obtained based on the traffic flow corresponding to the key time points contained in the node attributes of each traffic node; Extract the topology information of each traffic node, wherein the topology information includes the number of traffic nodes with which each traffic node has a corresponding relationship and the percentage change in traffic volume of the traffic nodes with which each traffic node has a corresponding relationship when the traffic volume of each traffic node increases by a percentage. The average percentage change in traffic flow of each traffic node is obtained by using the percentage change in traffic flow of its associated traffic nodes corresponding to the percentage increase in traffic flow of each traffic node. The average traffic flow at each traffic node is compared with a preset traffic flow reference value to obtain the ratio between the average traffic flow at each traffic node and the preset traffic flow reference value. The ratio between the average traffic flow of each traffic node and the preset traffic flow reference value is compared with the average percentage change of the associated traffic nodes corresponding to each traffic node; The traffic topology weight value of each traffic node is set based on the comparison between the ratio of the average traffic flow of each traffic node to the preset traffic flow reference value and the average percentage change of the associated traffic nodes corresponding to each traffic node, combined with the preset law enforcement management method information corresponding to each traffic node.

[0032] The technical effects of the above solution are as follows: By comprehensively considering the node attributes (such as traffic flow at key time points) and topology information (such as the number of associated nodes and the percentage change in traffic volume) of each traffic node, the solution can effectively improve the accuracy and rationality of the importance representation of each traffic node in the traffic network. By calculating the ratio of the average traffic flow to the preset traffic flow reference value, and the average percentage change of associated traffic nodes, the solution can effectively improve the matching between the weight value setting and the actual situation of the traffic node. Furthermore, when the actual state of a traffic node changes, the solution can effectively improve the sensitivity and accuracy of the traffic topology weight value of the traffic node in following the actual state change. On the other hand, the solution further improves the accuracy of the traffic topology weight value setting by leveraging the interrelationships between different nodes. Simultaneously, by utilizing the above solution, while increasing the number of operational parameters of the traffic nodes that need to be referenced, the solution can minimize the acquisition cycle and complexity of the traffic topology weight value, thereby increasing the efficiency of setting the traffic topology weight value while simultaneously increasing the number of operational parameters of the traffic nodes that need to be referenced.

[0033] Specifically, based on the comparison between the ratio of the average traffic flow of each traffic node to a preset traffic flow reference value and the average percentage change of associated traffic nodes corresponding to each traffic node, combined with the preset law enforcement management method information corresponding to each traffic node, the traffic topology weight value of each traffic node is set, including: When the ratio between the average traffic flow of each traffic node and the preset traffic flow reference value exceeds the average percentage change of the associated traffic nodes corresponding to each traffic node, the traffic node whose ratio between the average traffic flow of each traffic node and the preset traffic flow reference value exceeds the average percentage change of the associated traffic nodes corresponding to each traffic node is designated as the first target traffic node. Retrieve the traffic capacity and the number of branch roads corresponding to each first target traffic node; Extract the average traffic flow for each branch road contained in each first target traffic node; The average traffic flow of each branch road included in each first target traffic node is compared with the traffic capacity of the first target traffic node to obtain the traffic flow percentage value of each branch road included in each first target traffic node. The traffic flow percentage of each branch road is compared with the average percentage change of the associated traffic nodes corresponding to each first target traffic node. The branch roads whose traffic flow percentage is not lower than the average percentage change of the associated traffic nodes are extracted as target branch roads. The traffic topology weight value corresponding to the first target traffic node is obtained by combining the traffic flow percentage of the target intersection road with the preset topology weight benchmark value. The traffic topology weight value corresponding to the first target traffic node is obtained by the following formula:

[0034] Among them, Y 01 Y0 represents the traffic topology weight value corresponding to the first target traffic node; Y0 represents the preset topology weight benchmark value, wherein the topology weight benchmark value is obtained according to the actual application scenario requirements and experiments; B 01 This represents the average percentage change of associated traffic nodes corresponding to the first target traffic node; n represents the number of target branch roads corresponding to the first target traffic node; L 01i This represents the traffic flow percentage corresponding to the i-th target intersection road at the first target traffic node. Specifically, The traffic flow percentages of all n target branch roads corresponding to the first target traffic node are summed. This sum comprehensively considers the traffic flow contribution of all target branch roads, reflecting the overall contribution of the target branch roads to the traffic flow of the first target traffic node. Dividing the sum of all target branch road traffic flow percentages by the number of target branch roads, n, yields the average traffic flow percentage of each target branch road. This average average traffic flow percentage measures the average traffic flow contribution of each target branch road, making the calculation results more representative and stable. The average percentage change of associated traffic nodes corresponding to the first target traffic node is multiplied by the average traffic flow proportion of the target intersection road. This comprehensively considers the changes in surrounding traffic nodes and the average traffic flow contribution of the target intersection road, reflecting the combined effect of these two factors on the traffic topology weight value. Simultaneously, an adjustment coefficient is formed by adding the product to the initial value of 1. 1 represents maintaining the preset topology weight baseline value unchanged. The square root of the adjustment coefficient is used to further adjust the magnitude of weight changes, possibly to make the weight adjustment smoother and more reasonable, avoiding excessively large or small adjustments, and optimizing the weight calculation results according to actual needs. In the above formula, the traffic topology weight value Y corresponding to the first target traffic node is obtained by multiplying the preset topology weight baseline value Y0 by the adjustment coefficient obtained through a series of calculations. 01 Taking into account factors such as changes in associated traffic nodes and traffic flow at the target intersection, the preset benchmark value is dynamically adjusted based on actual traffic conditions to determine the weight of the first target traffic node in the traffic topology, which is then used for subsequent traffic management and other related decisions.

[0035] The technical effects of the above solution are as follows: By comparing the average traffic flow of traffic nodes with preset reference values, and the average percentage change of associated traffic nodes, the solution effectively improves the accuracy and efficiency of identifying traffic nodes with abnormal or critical traffic flow (i.e., the first target traffic node). Simultaneously, by analyzing the traffic flow proportion of each intersection road based on the traffic capacity and number of branch roads of the first target traffic node, the accuracy of the traffic topology weight value corresponding to each first target traffic node can be effectively improved. Furthermore, setting the traffic topology weight value for the first target traffic node based on the traffic flow proportion of the target intersection rather than all intersections within each traffic node effectively improves the efficiency of traffic topology weight value setting. Additionally, filtering intersections using the average percentage change of associated traffic nodes corresponding to the first target traffic node effectively improves the accuracy of target intersection road filtering, thereby further improving the matching between the traffic topology weight value setting and the actual traffic operation status of the first target traffic node, and enhancing the accuracy of the traffic topology weight value setting. Meanwhile, the above technical solutions and mathematical models only utilize parameter information of some branch roads of traffic nodes, which greatly reduces the utilization rate of branch roads. While ensuring that the weight value setting effectively improves the representation strength of traffic nodes, it minimizes the utilization rate of computing resources and improves energy efficiency.

[0036] Specifically, based on the comparison between the ratio of the average traffic flow of each traffic node to a preset traffic flow reference value and the average percentage change of associated traffic nodes corresponding to each traffic node, combined with the preset law enforcement management method information corresponding to each traffic node, the traffic topology weight value of each traffic node is set, which also includes: When the ratio between the average traffic flow of each traffic node and the preset traffic flow reference value does not exceed the average percentage change of the associated traffic nodes corresponding to each traffic node, the traffic node whose ratio between the average traffic flow of each traffic node and the preset traffic flow reference value does not exceed the average percentage change of the associated traffic nodes corresponding to each traffic node is taken as the second target traffic node. Extract the preset law enforcement management method information for each second target traffic node, wherein the law enforcement management method information includes the law enforcement resource distribution intensity coefficient corresponding to each fork road included in the second target traffic node, wherein the law enforcement resource distribution intensity is divided into strong, medium and weak; and the law enforcement resource distribution intensity coefficients corresponding to the strong, medium and weak levels of the law enforcement resource distribution intensity are 1.0, 0.8 and 0.5, respectively. Retrieve the traffic capacity and the number of branch roads corresponding to each second target traffic node; Extract the average traffic flow for each branch road contained in each second target traffic node; The traffic flow ratio of each branch road included in each second target traffic node is obtained by comparing the average traffic flow of each branch road with the traffic capacity of the second target traffic node. The traffic flow percentage of each branch road is compared with the average percentage change of the associated traffic nodes corresponding to each second target traffic node. The branch roads whose traffic flow percentage is not lower than the average percentage change of the associated traffic nodes are extracted as target branch roads. The traffic topology weight value corresponding to the second target traffic node is obtained by combining the traffic flow ratio corresponding to the target intersection road with the law enforcement resource distribution intensity coefficient corresponding to the target intersection road and the preset topology weight benchmark value. The traffic topology weight value corresponding to the second target traffic node is obtained by the following formula:

[0037] Among them, Y 02 Y0 represents the traffic topology weight value corresponding to the second target traffic node; B represents the preset topology weight baseline value. 02This represents the average percentage change of associated traffic nodes corresponding to the second target traffic node; m represents the number of target branch roads corresponding to the second target traffic node; L 02i This represents the traffic flow percentage corresponding to the i-th target intersection road at the second target traffic node; k i This represents the intensity coefficient of law enforcement resource distribution corresponding to the i-th target intersection road at the second target traffic node. Specifically, The exponential function part is based on the law enforcement resource distribution intensity coefficient k. i The value of k changes. i When the value is relatively large (high intensity of law enforcement resource distribution), exp(-k) i When k approaches 0, the denominator approaches 1; i When the value is small, the denominator changes accordingly. The numerator B 02 This is the average percentage change of associated traffic nodes. This fraction comprehensively considers the changes in associated traffic nodes and the intensity of law enforcement resource distribution, combining the two through a functional relationship to measure the combined impact of these two factors on the weight calculation. It is the intensity coefficient k of law enforcement resource distribution corresponding to the i-th target fork road. i Traffic flow percentage L corresponding to the intersection 02i The multiplication factor takes into account both the allocation of law enforcement resources and the contribution of traffic flow, reflecting the combined impact of the two factors on the traffic topology weight value. The differences mentioned above for all m target branch roads corresponding to the second target traffic node are summed. Taking into account changes in associated traffic nodes, distribution of law enforcement resources, and traffic flow across all target branch roads, the total combined influence of these factors on all target branch roads is obtained, reflecting the overall effect of the target branch roads on the traffic topology weight value. Dividing this sum by the number of target branch roads, m, yields the average value of the combined influence of these factors, making the result more representative and stable. This value is used to measure the average influence of each target branch road on the traffic topology weight under the combined effect of these factors.

[0038] The technical effects of the above solution are as follows: By comparing the average traffic flow of traffic nodes with preset reference values, and the average percentage change of associated traffic nodes, the solution effectively improves the accuracy and efficiency of identifying traffic nodes with abnormal or critical traffic flow (i.e., the second target traffic node). Simultaneously, by analyzing the traffic flow proportion of each intersection road based on the traffic capacity and number of branch roads of the second target traffic node, the accuracy of the traffic topology weight value corresponding to each second target traffic node can be effectively improved. Furthermore, setting the traffic topology weight value for the second target traffic node based on the traffic flow proportion of the target intersection rather than all intersections within each traffic node effectively improves the efficiency of traffic topology weight value setting. Additionally, filtering intersections using the average percentage change of associated traffic nodes corresponding to the second target traffic node effectively improves the accuracy of target intersection road filtering, thereby further improving the matching between the traffic topology weight value setting and the actual traffic operation status of the second target traffic node, and enhancing the accuracy of the traffic topology weight value setting. Meanwhile, the above technical solutions and mathematical models only utilize parameter information of some branch roads of traffic nodes, which greatly reduces the utilization rate of branch roads. While ensuring that the weight value setting effectively improves the representation strength of traffic nodes, it minimizes the utilization rate of computing resources and improves energy efficiency.

[0039] On the other hand, in the aforementioned technical solution, adding preset enforcement management information during the weight value setting process for the second target traffic node where the ratio between the average traffic flow and the preset traffic flow reference value does not exceed the average percentage change of the associated traffic nodes corresponding to each traffic node can effectively improve the rationality of the weight value setting for the second target traffic node and the accuracy of the representation of the importance of the second target traffic node. This prevents the problem of unreasonable topology weight value setting for the second target traffic node when traffic flow anomalies occur, as the ratio between the average traffic flow and the preset traffic flow reference value does not exceed the average percentage change of the associated traffic nodes corresponding to each traffic node. This further improves the accuracy of the weight value setting for the second target traffic node.

[0040] In this embodiment, the law enforcement scene monitoring module performs law enforcement scene identification, specifically including: Extract video stream data from traffic data in law enforcement scenarios and preprocess it to obtain preprocessed video stream data; Feature extraction is performed on the preprocessed video stream data, including extracting visual features such as vehicle shape, license plate, and traffic signs from traffic camera video data, extracting motion features such as speed, acceleration, and driving direction from vehicle motion data, extracting acoustic features such as abnormal sounds and vehicle collision sounds from audio and video data, and extracting relevant features such as law enforcement event type and processing result from law enforcement record data. Based on the extraction results, an initial feature set of traffic data is obtained. Key features relevant to law enforcement scenarios are extracted from the initial feature set and integrated into a subset of key features. In the overload scenario, key features include vehicle load, cargo volume, axle load, etc.; in the "two passengers, one dangerous goods and one heavy" scenario, key features include vehicle operation qualifications, driving route, dangerous goods markings, etc.; in the illegal operation scenario, key features include passenger boarding and alighting frequency, operating time, vehicle operation permit information, etc. Acquire information on the topology of the traffic network and the preset law enforcement management methods (such as the key areas of law enforcement in different regions and the distribution of law enforcement resources), as well as the node attributes of each traffic node (such as intersections and key locations on road sections) (such as traffic flow level and accident frequency level). Based on the node attributes and topology information of each traffic node, as well as the preset law enforcement management method information, the traffic topology weight value of each traffic node is determined; for example, high weights are assigned to accident-prone intersections, and corresponding weights are assigned to road sections connecting important locations, so as to reflect the importance of different nodes in law enforcement scenarios. Based on the traffic topology weight value of each traffic node, the basic values ​​of relevant indicators for each law enforcement scenario in the traffic data are calculated. The basic values ​​of normal traffic flow and vehicle speed for different road sections at different times are calculated as reference standards for judging abnormal situations. Obtain the temporal feature information corresponding to traffic data in law enforcement scenarios, such as changes in traffic flow and fluctuations in vehicle speed over time. Based on the temporal feature information, extract the temporal sequence data of relevant indicators (such as vehicle throughput and number of violations) for each law enforcement scenario from the traffic data in law enforcement scenarios to reflect their changing patterns over time. Based on the time series data of relevant indicators for each law enforcement scenario, the law enforcement scenario characteristic information of that indicator is determined. By analyzing the time series data of traffic flow, its peak and off-peak period characteristics, the frequency and duration of congestion, etc. are determined. From the time series data of the number of violations, the high-incidence time periods and regional characteristics of violations are determined. The time series data and basic values ​​of each law enforcement scenario-related indicator are used as input samples for the model, and the law enforcement scenario feature information of each indicator is used as output samples for the model. The preset network model is trained to obtain the recognition model of each law enforcement scenario-related indicator. The identification model based on the relevant indicators of each law enforcement scenario identifies the target features in the video stream data, determines the current law enforcement scenario type, and sets the corresponding law enforcement scenario label, such as a suspected overloading scenario, a violation scenario of "two passengers, one dangerous goods and one heavy" or a suspected illegal operation scenario, etc. In this embodiment, the law enforcement scene monitoring module for law enforcement scene identification further includes: By using the identification model of relevant indicators for each law enforcement scenario, the target law enforcement scenario features corresponding to the target time series data of each key feature in the key feature subset are obtained; Based on the target law enforcement scenario features of each key feature, obtain the first law enforcement scenario feature for each law enforcement scenario; Obtain the changes in the target law enforcement scenario characteristics of each law enforcement scenario in the traffic data of the law enforcement scenario, and determine the law enforcement scenario change rules for each law enforcement scenario based on the changes. Enforcement scenarios with a similarity to the rule of change of enforcement scenarios greater than or equal to a preset threshold are identified as the same type of enforcement scenario. The second target enforcement scenario feature of any enforcement scenario in each type of enforcement scenario is identified as the final enforcement scenario feature of that type of enforcement scenario. In this embodiment, the time-series data of vehicle load is analyzed by an identification model to obtain the overload risk characteristics at different time periods; the time-series data of vehicle travel routes is analyzed to obtain abnormal characteristics of deviation from the prescribed routes, etc. By combining the key characteristics of vehicle load, travel route, and operating qualifications, the target law enforcement scenario characteristics are determined to identify the preliminary feature description of the law enforcement scenario to which a certain vehicle belongs (such as a scenario suspected of being overloaded, a "two-passenger-one-dangerous-one-heavy" violation scenario, or a scenario suspected of illegal operation, etc.). The variation patterns of traffic flow in different seasons, weekdays and rest days, as well as the changing trend of the frequency of violations over time, are analyzed to summarize the variation patterns of different law enforcement scenarios in time and space. Multiple road segment scenarios with similar traffic flow variation patterns, violation types and frequencies are grouped into one category. The key law enforcement features of one typical scenario (such as specific violation combination patterns, correlation characteristics between traffic congestion and violations, etc.) are selected as the final features of this type of law enforcement scenario for subsequent law enforcement scenario judgment and law enforcement decision support.

[0041] In this embodiment, determining whether a vehicle in a law enforcement scenario is abnormal specifically includes: When the axle weight of a vehicle exceeds the vehicle axle weight threshold in the real-time traffic data of law enforcement scenarios, and the change in vehicle height exceeds the range of vehicle height change, combined with abnormal sounds in audio and video data, such as abnormal tire friction sounds, abnormal noises caused by excessive pressure on the vehicle frame, it is judged that there may be overloading behavior. When the frequency of passenger boarding and alighting exceeds the passenger boarding and alighting frequency threshold and the operating time exceeds the operating time range, the voice data of audio and video data is analyzed to determine whether the conversation between the passenger and the driver involves operation-related information. If there is evidence to support illegal operation, it is judged as suspected illegal operation. If the vehicle's operating qualification markings do not conform to the vehicle operating qualification compliance markings, or if the driving route coordinates are not within the range of the prescribed driving route coordinate set, such as whether the vehicle body has markings that do not conform to the operating qualification, or whether it is carrying passengers or goods that do not conform to the nature of the vehicle's operation; check whether the vehicle deliberately avoids checkpoints or drives in prohibited areas during the driving process, then it is determined that the "two-passenger-one-dangerous-one-heavy" vehicle is in violation. Once an anomaly is detected in a vehicle within a law enforcement scenario, an early warning mechanism is immediately triggered. Based on the type of law enforcement scenario and the category of anomaly, an early warning message is generated by combining audio and video data from the law enforcement process. For example, the early warning message may include video screenshots of the abnormal vehicle and key audio clips, so that law enforcement officers can more intuitively understand the abnormal situation of the vehicle and respond quickly.

[0042] In this embodiment, by real-time monitoring of vehicle axle load, changes in vehicle height, and passenger boarding and alighting frequency, combined with audio and video analysis, the system effectively identifies overloading, illegal operation, and violations by passenger vehicles, hazardous material transport vehicles, and heavy-duty vehicles, triggering an early warning mechanism in a timely manner. This improves enforcement efficiency, helps ensure road transport safety, maintains market order, and provides law enforcement personnel with intuitive and convenient evidence for enforcement.

[0043] In this embodiment, the traffic enforcement situation model specifically includes: Monitor the temporal and spatial distribution of traffic data in law enforcement scenarios and obtain detection results, such as the distribution of detection results under different road sections and different time periods; Based on the detection results, cluster analysis is performed on data such as traffic flow and frequency of violations in different law enforcement scenarios to form different situation categories, determine the clustering characteristics of various law enforcement scenarios, and the changing direction and degree trend of each situation category; The correlation between traffic violations and traffic flow, weather conditions, time, and accident information is determined. Based on the correlation, a visualized traffic enforcement situation map is generated, thereby constructing a traffic enforcement situation model. In the traffic enforcement situation map, different road segments are visualized and marked according to their traffic enforcement situation categories. For example, red indicates high violation risk areas, green indicates low violation risk areas, and icons of different sizes represent traffic flow. At the same time, weather information (such as using weather icons to represent temperature, precipitation, etc.), time information (distinguished by time axis or different time periods by color), and accident location and type (displayed in the situation map through annotations) are overlaid on the traffic enforcement situation map.

[0044] In this embodiment, the traffic enforcement situation model monitors and analyzes traffic data in enforcement scenarios, enabling the spatiotemporal distribution clustering of traffic flow and frequency of violations to form situation categories and reveal their changing trends. It generates an intuitive traffic enforcement situation map, which visually displays the risk level and traffic volume of different road sections, as well as weather, time, and accident information through colors and icons. This provides law enforcement personnel with a comprehensive and dynamic analysis of the enforcement situation, helping to optimize the allocation of enforcement resources and improve enforcement efficiency.

[0045] In this embodiment, the classification, organization, and analysis of feedback information from law enforcement personnel also includes: Text information analysis is performed on the feedback information from law enforcement personnel, and the text information is segmented into multi-level topic trees; The text information is segmented into words, and similarity matching is performed based on the constructed thesaurus of synonyms in the field of traffic law enforcement and each extracted word; Based on the similarity matching results, duplicates are removed to obtain the target word set, and keywords are extracted from the target word set. Based on the keyword extraction results from the relevant feedback information clusters of each law enforcement scenario, determine the keywords included in each layer of the topic tree; Different categories are distinguished: feedback on the accuracy of law enforcement scenario judgments, feedback on the effectiveness of law enforcement decision implementation, and suggestions for system function improvement.

[0046] In this embodiment, the first layer of the theme tree can be classified according to law enforcement scenarios (overloading, passenger vehicles, dangerous goods vehicles, heavy vehicles, illegal operations, etc.), the second layer can be classified according to feedback type (judgment accuracy, decision execution effect, system function improvement, etc.), and the third layer is specific target keywords. The sorted and analyzed keyword information is presented to relevant personnel in an intuitive way, so that they can understand the key content of the feedback from law enforcement personnel and provide a basis for optimizing the law enforcement scenario monitoring model, situational awareness analysis algorithm and law enforcement decision support strategy.

[0047] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing, characterized in that: include: The law enforcement scene monitoring module is configured to acquire traffic data in law enforcement scenes in real time, extract vehicle feature data, record audio and video data, and perform law enforcement scene recognition, determine feature parameters, and judge whether the vehicles in the law enforcement scene are abnormal. Among them, the law enforcement scenario recognition is configured to analyze the node attributes and topology of each traffic node, determine the traffic topology weight value of each traffic node in combination with the preset law enforcement management method, determine the law enforcement scenario feature information based on the traffic topology weight value combined with time series features and sequence data, identify target features to determine the scenario type and set labels.

2. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 1, characterized in that, The law enforcement scenario monitoring module performs law enforcement scenario identification, specifically including: Extract video stream data from traffic data in law enforcement scenarios and preprocess it to obtain preprocessed video stream data; Feature extraction is performed on the preprocessed video stream data, and an initial feature set for traffic data is obtained based on the extraction results; Extract key features relevant to the law enforcement scenario from the initial feature set and integrate them into a subset of key features; Obtain information on the topology of the traffic network, pre-defined law enforcement management methods, and the node attributes of each traffic node; Based on the node attributes and topology information of each traffic node, as well as the preset law enforcement management method information, determine the traffic topology weight value of each traffic node; Based on the traffic topology weight value of each traffic node, the basic values ​​of the relevant indicators for each law enforcement scenario in the traffic data are calculated. Obtain the temporal feature information corresponding to traffic data in law enforcement scenarios, and extract the temporal sequence data of relevant indicators for each law enforcement scenario from the traffic data based on the temporal feature information; The law enforcement scenario characteristic information of each indicator is determined based on the time series data of the relevant indicators for each law enforcement scenario; The time series data and basic values ​​of each law enforcement scenario-related indicator are used as input samples for the model, and the law enforcement scenario feature information of each indicator is used as output samples for the model. The preset network model is trained to obtain the recognition model of each law enforcement scenario-related indicator. The identification model, based on the relevant indicators of each law enforcement scenario, identifies target features in the video stream data, determines the current law enforcement scenario type, and sets corresponding law enforcement scenario labels.

3. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 2, characterized in that, Based on the node attributes and topology information of each traffic node, as well as the preset law enforcement management method information, the traffic topology weight value of each traffic node is determined, including: Extract the traffic flow corresponding to the key time nodes contained in the node attributes of each traffic node; The average traffic flow is obtained based on the traffic flow corresponding to the key time points contained in the node attributes of each traffic node; Extract the topology information of each traffic node, wherein the topology information includes the number of traffic nodes with which each traffic node has a corresponding relationship and the percentage change in traffic volume of the traffic nodes with which each traffic node has a corresponding relationship when the traffic volume of each traffic node increases by a percentage. The average percentage change in traffic flow of each traffic node is obtained by using the percentage change in traffic flow of its associated traffic nodes corresponding to the percentage increase in traffic flow of each traffic node. The average traffic flow at each traffic node is compared with a preset traffic flow reference value to obtain the ratio between the average traffic flow at each traffic node and the preset traffic flow reference value. The ratio between the average traffic flow of each traffic node and the preset traffic flow reference value is compared with the average percentage change of the associated traffic nodes corresponding to each traffic node; The traffic topology weight value of each traffic node is set based on the comparison between the ratio of the average traffic flow of each traffic node to the preset traffic flow reference value and the average percentage change of the associated traffic nodes corresponding to each traffic node, combined with the preset law enforcement management method information corresponding to each traffic node.

4. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 3, characterized in that, Based on the comparison between the ratio of the average traffic flow of each traffic node to a preset traffic flow reference value and the average percentage change of associated traffic nodes corresponding to each traffic node, combined with the preset law enforcement management method information corresponding to each traffic node, the traffic topology weight value of each traffic node is set, including: When the ratio between the average traffic flow of each traffic node and the preset traffic flow reference value exceeds the average percentage change of the associated traffic nodes corresponding to each traffic node, the traffic node whose ratio between the average traffic flow of each traffic node and the preset traffic flow reference value exceeds the average percentage change of the associated traffic nodes corresponding to each traffic node is designated as the first target traffic node. Retrieve the traffic capacity and the number of branch roads corresponding to each first target traffic node; Extract the average traffic flow for each branch road contained in each first target traffic node; The average traffic flow of each branch road included in each first target traffic node is compared with the traffic capacity of the first target traffic node to obtain the traffic flow percentage value of each branch road included in each first target traffic node. The traffic flow percentage of each branch road is compared with the average percentage change of the associated traffic nodes corresponding to each first target traffic node. The branch roads whose traffic flow percentage is not lower than the average percentage change of the associated traffic nodes are extracted as target branch roads. The traffic topology weight value corresponding to the first target traffic node is obtained by combining the traffic flow percentage of the target intersection road with the preset topology weight benchmark value. When the ratio between the average traffic flow of each traffic node and the preset traffic flow reference value does not exceed the average percentage change of the associated traffic nodes corresponding to each traffic node, the traffic node whose ratio between the average traffic flow of each traffic node and the preset traffic flow reference value does not exceed the average percentage change of the associated traffic nodes corresponding to each traffic node is taken as the second target traffic node. Extract the preset law enforcement management method information for each second target traffic node, wherein the law enforcement management method information includes the law enforcement resource distribution intensity coefficient corresponding to each fork road included in the second target traffic node, wherein the law enforcement resource distribution intensity is divided into strong, medium and weak; and the law enforcement resource distribution intensity coefficients corresponding to the strong, medium and weak levels of the law enforcement resource distribution intensity are 1.0, 0.8 and 0.5, respectively. Retrieve the traffic capacity and the number of branch roads corresponding to each second target traffic node; Extract the average traffic flow for each branch road contained in each second target traffic node; The traffic flow ratio of each branch road included in each second target traffic node is obtained by comparing the average traffic flow of each branch road with the traffic capacity of the second target traffic node. The traffic flow percentage of each branch road is compared with the average percentage change of the associated traffic nodes corresponding to each second target traffic node. The branch roads whose traffic flow percentage is not lower than the average percentage change of the associated traffic nodes are extracted as target branch roads. The traffic topology weight value corresponding to the second target traffic node is obtained by combining the traffic flow ratio corresponding to the target intersection road with the law enforcement resource distribution intensity coefficient corresponding to the target intersection road and the preset topology weight benchmark value.

5. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 3, characterized in that, The law enforcement scene monitoring module, which performs law enforcement scene identification, also includes: By using the identification model of relevant indicators for each law enforcement scenario, the target law enforcement scenario features corresponding to the target time series data of each key feature in the key feature subset are obtained; Based on the target law enforcement scenario features of each key feature, obtain the first law enforcement scenario feature for each law enforcement scenario; Obtain the changes in the target law enforcement scenario characteristics of each law enforcement scenario in the traffic data of the law enforcement scenario, and determine the law enforcement scenario change rules for each law enforcement scenario based on the changes. Enforcement scenarios with a similarity to the rule of change of enforcement scenarios greater than or equal to a preset threshold are identified as the same type of enforcement scenario. The second target enforcement scenario feature of any enforcement scenario in each type of enforcement scenario is identified as the final enforcement scenario feature of that type of enforcement scenario.

6. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 2, characterized in that, The feature parameters are determined based on the law enforcement scenario identification results, which include: overloaded and oversized scenarios, illegal operation scenarios, and "two types of passenger vehicles, one type of dangerous goods vehicle, and one type of heavy vehicle" scenarios. Based on the overloaded and oversized scenario, for truck-type vehicles, feature parameters such as axle group type, number of axles, weighing data, and length, width, and height data are determined. Based on the illegal operation scenario, feature parameters such as operating license, business scope, and validity period of the license are determined. Based on the "two types of passenger vehicles, one type of dangerous goods vehicle, and one type of heavy vehicle" scenario, parameter data such as vehicle operation qualification compliance mark and the set of coordinates of the prescribed driving route are determined.

7. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 6, characterized in that, Determining whether a vehicle is abnormal in a law enforcement scenario specifically includes: When the axle weight of a vehicle exceeds the vehicle axle weight threshold in the real-time traffic data of law enforcement scenarios, and the change in vehicle height exceeds the range of vehicle height change, combined with abnormal sounds in audio and video data, it is determined that there may be overloading behavior. When the frequency of passenger boarding and alighting exceeds the passenger boarding and alighting frequency threshold and the operating time exceeds the operating time range, the voice data of audio and video data is analyzed to determine whether the conversation between the passenger and the driver involves operation-related information. If there is evidence to support illegal operation, it is judged as suspected illegal operation. If the vehicle operation qualification mark does not conform to the vehicle operation qualification compliance mark, or the driving route coordinates are not within the range of the prescribed driving route coordinate set, then the "two passenger vehicles, one dangerous goods vehicle and one heavy vehicle" are judged to be in violation. Once an anomaly is detected in a vehicle within a law enforcement scenario, an early warning mechanism is immediately triggered, generating warning information based on the type of law enforcement scenario and the category of anomaly, combined with audio and video data from the law enforcement process.

8. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 1, characterized in that, Also includes: The situational awareness analysis module is configured to analyze the traffic law enforcement situation based on the output of the law enforcement scenario monitoring module and using a traffic law enforcement situational model. The law enforcement decision support module is configured to optimize the allocation of law enforcement resources based on the results of traffic law enforcement situation analysis, predict violation trends and key areas based on historical and real-time data, and formulate law enforcement plans. The law enforcement feedback interaction module is configured to receive feedback information from law enforcement personnel, classify and analyze it, and push law enforcement scenario information, early warning information and law enforcement plans to law enforcement personnel in real time.

9. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 8, characterized in that, The traffic enforcement situation model specifically includes: Monitor the temporal and spatial distribution of traffic data in law enforcement scenarios and obtain detection results; Based on the detection results, cluster analysis is performed on data from different law enforcement scenarios to form different situation categories, determine the clustering characteristics of various law enforcement scenarios, and the changing direction and degree trend of each situation category; The correlation between traffic violations and traffic flow, weather conditions, time, and accident information is determined, and a visualized traffic enforcement situation map is generated based on the correlation, thereby constructing a traffic enforcement situation model. The traffic enforcement situation map visually marks different road sections according to their traffic enforcement situation categories, and uses icons of different sizes to represent the traffic flow. At the same time, weather information, time information, and accident location and type are overlaid on the traffic enforcement situation map.

10. The traffic integrated law enforcement situational awareness and analysis system based on big data and cloud computing as described in claim 8, characterized in that, The process of categorizing, organizing, and analyzing feedback from law enforcement personnel also includes: Text information analysis is performed on the feedback information from law enforcement personnel, and the text information is segmented into multi-level topic trees; The text information is segmented into words, and similarity matching is performed based on the constructed thesaurus of synonyms in the field of traffic law enforcement and each extracted word; Based on the similarity matching results, duplicates are removed to obtain the target word set, and keywords are extracted from the target word set. Based on the keyword extraction results from the relevant feedback information clusters of each law enforcement scenario, determine the keywords included in each layer of the topic tree; Different categories are distinguished: feedback on the accuracy of law enforcement scenario judgments, feedback on the effectiveness of law enforcement decision implementation, and suggestions for system function improvement.

Images