Traffic analysis cloud platform and method based on big data, and storage medium

Abstract

The invention discloses a traffic analysis cloud platform and method based on big data and a storage medium. The traffic analysis cloud platform comprises a basic traffic data acquisition / access module, a traffic data management module, a traffic data processing module, a big data analysis module, a visualization and business application module, a data security management module and a platform basic service module. The traffic analysis cloud platform based on big data constructs a complete technical chain of collection-storage-analysis-application so as to realize real-time perception and intelligent decision making, refined analysis and accurate service and cross-system cooperation and emergency response. The invention discloses a traffic analysis method based on big data. The traffic analysis method comprises the steps of data acquisition and transmission, data storage and management, data analysis and processing, data application and feedback and security and expansibility of a cloud platform. The cloud platform and the analysis method can be applied to multiple fields of urban traffic management, can improve traffic operation efficiency and management efficiency, and have the advantages of real-time performance, accuracy and the like.

Description

Technical Field

[0001] The present invention belongs to the field of traffic management technology, and specifically relates to a traffic analysis cloud platform, method and storage medium based on big data. Background Art

[0002] With the acceleration of intelligent urban transportation, the application of traffic big data analysis platforms in areas such as urban traffic management and public service optimization is becoming increasingly important. Existing technologies have already implemented a collaborative architecture of "small data centers + edge computing + AI + the Internet of Things," achieving progress in real-time data processing and multi-sector coverage (such as highways, urban transportation, and port logistics). The market is characterized by leading companies with a full-stack presence, while smaller and medium-sized vendors focus on specific scenarios.

[0003] However, existing technologies still have significant shortcomings: Equipment and system standards vary from manufacturer to manufacturer, making cross-segment and cross-platform data integration difficult. This leads to widespread "siloed" architectures, hindering cross-regional data sharing and collaborative management. For example, due to differences in construction timelines and standards, traditional transportation systems cannot effectively support cross-segment dispatching, toll collection, and other collaborative tasks. Initial investments in sensor networks, communication towers, and data centers are substantial, and these costs are difficult to amortize before large-scale deployment. Traditional systems also have numerous failure points and low operational and maintenance efficiency, resulting in a long payback period. Vehicle-road systems rely on real-time data transmission and are vulnerable to cyberattacks. Although pilot programs have begun for technologies like quantum encryption, full implementation is still time-consuming, and some platforms still face data leakage risks. Edge-cloud collaboration and multimodal data fusion require complex technologies, and small and medium-sized enterprises lack sufficient technical expertise. Furthermore, significant regional variations in technology application levels make standardized solutions difficult to develop. Some platforms opt for on-premises deployments to ensure security, hindering rapid rollout and hindering data sharing and cloud-based data analysis, making them unable to adapt to the dynamic demands of large-scale transportation networks. In addition, when dealing with specific scenarios such as dynamic regulation of traffic congestion, cross-departmental emergency response, and logistics efficiency optimization, existing technologies are unable to achieve efficient decision-making support due to insufficient data integration capabilities and limited algorithm model accuracy. Summary of the Invention

[0004] The purpose of the present invention is to provide a traffic analysis cloud platform, method and storage medium based on big data, which can solve the technical problems of insufficient data integration capability and low operation and maintenance efficiency in the existing technology.

[0005] To achieve the above-mentioned purpose, a specific embodiment of the present invention provides a traffic analysis cloud platform based on big data, which includes: a basic traffic data collection / access module, a traffic data management module, a traffic data processing module, a big data analysis module, and a visualization and business application module; wherein, The basic traffic data collection / access module is used to collect, classify and store multi-source traffic data; The traffic data management module is connected to the basic traffic data collection / access module, and the traffic data management module is used to set the data specifications and storage location for collecting multi-source traffic; The traffic data processing module is connected to the traffic data management module, and the traffic data processing module is used to pre-process the data; The big data analysis module is connected to the traffic data processing module, and the big data analysis module is used to perform basic analysis on the pre-processed data and generate analysis results; The visualization and business application module is connected to the big data analysis module, and the visualization and business application module is used to display analysis results and provide business applications.

[0006] In one or more embodiments of the present invention, the basic traffic data collection / access module is used to collect basic information data related to traffic and upload it to various interfaces of the cloud platform. The basic traffic data includes static data, dynamic data and user behavior data, wherein: The static data includes road network topology, POI and transportation facility locations; The dynamic data includes GPS tracks, ETC gantry records, video surveillance, geomagnetic coils and meteorological sensor records; The user behavior data includes bus / subway card swiping records, navigation APP requests, and shared travel order data.

[0007] In one or more embodiments of the present invention, the preprocessing of the traffic data processing module includes data warehouse planning, data cleaning, data filling and data integration of basic traffic data, wherein data cleaning includes data denoising for filtering drift points in the data, the data filling includes filling missing sensor values using a difference algorithm, and the data integration includes unifying data of different frequencies to the same timestamp and coordinate system.

[0008] In one or more embodiments of the present invention, the big data analysis module includes: a basic analysis unit, a machine learning model unit, a graph computing application unit, a stream processing engine unit, and an edge computing optimization unit, wherein: The basic analysis unit is used to count the traffic volume and average speed of road sections by time window, mine commuting patterns through subway card swipe data, and classify congestion levels based on speed thresholds or congestion indexes; Machine learning model units, including a short-term traffic condition prediction model based on LSTM / Transformer, a traffic accident identification model based on CNN, a ride-hailing dispatch strategy optimization model based on reinforcement learning, and a traffic policy simulation model based on SUMO+AI Agent; The graph computing application unit uses Neo4j to identify key nodes in the road network and uses GraphSAGE to simulate congestion propagation paths; The stream processing engine unit uses Flink for stateful computing and combines it with Kafka for event distribution. The edge computing optimization unit deploys the YOLOv5 model on the NVIDIA Jetson edge device to achieve real-time analysis of video streams and control the latency within the preset value.

[0009] In one or more embodiments of the present invention, the visualization and business application module includes a visualization tool chain unit and a business system integration unit, wherein the visualization tool chain unit includes: a basic display component, a spatiotemporal interaction component, and a three-dimensional simulation component, wherein, The basic display component uses Tableau / Power BI to generate a traffic flow heat map, which presents the statistical results of traffic data in an intuitive two-dimensional graphic, focusing on traffic distribution and trend characteristics; The spatiotemporal interaction component uses Kepler.gl to draw a vehicle trajectory flow diagram. The spatiotemporal interaction component analyzes the dynamic correlation of traffic data based on the spatiotemporal dimension and mines behavioral patterns and spatial interaction features. The three-dimensional simulation component constructs a digital twin road network through CesiumJS, and the digital twin road network is used to realize three-dimensional visualization and dynamic simulation deduction of traffic scenes.

[0010] In one or more embodiments of the present invention, the business system integration unit includes a signal control interface, an emergency command interface, and a public service interface, wherein: The signal control interface connects to the SCATS signal system through the API and issues optimization solutions; The emergency command interface pushes the accident prediction results to the traffic police command screen, wherein the response time of pushing the prediction results to the traffic police command screen is lower than a preset response time threshold; The public service interface publishes real-time traffic conditions through the AutoNavi API to guide users to avoid congestion.

[0011] In one or more embodiments of the present invention, a data security management module is also included. The data security management module is connected to the aforementioned modules and is used to ensure the data security of the cloud platform. The functions of the data security management module include access control, data encryption, data backup and recovery, security auditing, data desensitization, security monitoring and early warning; the data security management module is also used to reasonably divide and isolate the internal network of the cloud platform.

[0012] In one or more embodiments of the present invention, a platform basic service module is also included, which is connected to the aforementioned modules. The platform basic service module provides basic services for the entire cloud platform. The basic services include at least database middleware, TCP service framework, authentication center, service center, unified monitoring, configuration center, message queue, task scheduling platform, distributed cache, file service, log platform, development interface platform, distributed deployment platform and development API gateway.

[0013] In another aspect of the present invention, a traffic analysis method based on big data is provided. The traffic analysis method is applied to the above-mentioned traffic analysis cloud platform, and the traffic analysis method includes the following steps: Collect sensor data through IoT protocols, use distributed message queues to process high-concurrency data streams, and simultaneously capture Internet traffic data; Store structured data in a distributed database or time series database, and unstructured data in an object storage service. Use a metadata management system to optimize data classification and retrieval. Use stream processing engines for real-time data cleaning and statistics, combined with machine learning models for road condition prediction and accident detection; Generate traffic situation visualization results through visualization tools and connect to business systems to achieve dynamic control; Encryption technology is used to ensure data transmission and storage security, fine-grained access control is implemented based on the permission control model, and traffic peaks are handled through containerized deployment and elastic expansion mechanisms.

[0014] In another aspect of the present invention, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, the steps of a traffic analysis cloud platform based on big data are implemented.

[0015] Beneficial effects: Compared with the prior art, the present invention has the following significant improvements: The big data-based traffic analysis cloud platform in this invention utilizes a hierarchical design consisting of a "basic traffic data collection / access module—traffic data management module—traffic data processing module—big data analysis module—visualization and business application module," creating a complete "collection-storage-analysis-application" technology chain. The basic traffic data collection module accesses multi-source data from traffic checkpoints, electronic police, and other sources in real time. After cleaning and integration by the data management and processing module, the big data analysis module uses an LSTM / Transformer model to predict road conditions for the next 30 minutes (MAPE error <15%). Finally, dynamic control is achieved through the visualization module's CesiumJS digital twin road network or SCATS signal system interface, forming an automated closed-loop from data collection to decision-making and execution, improving response efficiency by over 60% compared to traditional systems.

[0016] This system utilizes a distributed "core cloud + edge node" architecture. Edge nodes utilize NVIDIA Jetson devices running the YOLOv5 algorithm to analyze video streams with 100ms latency. The core cloud leverages XEBS distributed storage to support elastic scaling of petabyte-level data, addressing the "data silos" and high hardware investment challenges of traditional systems. Standardized data interfaces integrate data from nine major transportation sectors and connect to third-party systems (such as AutoNavi's real-time traffic data) through open APIs, breaking down barriers to cross-departmental collaboration. In a pilot application in Shanghai, cross-regional data sharing reduced vehicle congestion by 15% and increased average vehicle speeds by 10%.

[0017] The big data analysis module in this invention combines a CNN model (with >90% accuracy in identifying traffic accidents) with GraphSAGE graph computation to identify key nodes in road networks and predict congestion propagation paths. Application of this model in Baotou's central urban area has increased the proactive detection rate of traffic incidents by 53% and shortened response times by 18%. Reinforcement learning (DQN) dynamically optimizes ride-hailing dispatch strategies, reducing idle driving rates by 20%. Simulations using SUMO and AI Agents can also be used to deduce the impact of policies such as odd-even license plate restrictions, providing a scientific basis for traffic planning.

[0018] The visualization toolchain component of this invention transforms complex traffic data into intuitive graphics using Tableau heat maps, Kepler.gl trajectory maps, and CesiumJS 3D simulation. This allows managers to monitor traffic flow in real time in a simulated scenario in Shenzhen's Futian CBD. Business system integration achieves a trinity of "signal control, emergency command, and public services": Traffic light optimization solutions are distributed to the SCATS system via an API, accident prediction results are pushed to traffic police screens within one minute, and AutoNavi's API guides users to avoid traffic jams. As a result, traffic efficiency in specific directions at the Chongqing Sikmli Interchange during the morning rush hour increased by 63.8%.

[0019] The data security management module of this invention utilizes triple-copy storage, HTTPS encryption, and quantum cryptography to achieve 99.99999% reliability during platform use. Applications and predictions have shown that the risk of data leakage in Guizhou's highway maintenance system applications has been reduced by 80%. The platform's basic service module, deployed in a Kubernetes containerized manner, supports dynamic scaling of computing resources. During peak traffic periods such as the National Day holiday, Chongqing's expressway traffic speeds increased by 5.4%. Furthermore, hardware reuse and automated operations and maintenance have reduced IT costs by 40%, achieving both security and cost-effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments described in the present invention. Those skilled in the art can also derive other drawings based on these drawings without inventive effort.

[0021] Figure 1 This is a schematic diagram of the structure of a traffic analysis cloud platform based on big data in one embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of a big data analysis module in one embodiment of the present invention; Figure 3 A schematic diagram of the structure of a deep visualization and business application module in one embodiment of the present invention; Figure 4 Flowchart of a traffic analysis method based on big data in one embodiment of the present invention; Figure 5 This is a diagram of the traffic analysis cloud platform application architecture in one embodiment of the present invention; Figure 6 This is a demonstration diagram of the application of the traffic analysis cloud platform in one embodiment of the present invention; Figure 7 This is a diagram showing the application results of the traffic analysis cloud platform in one embodiment of the present invention. DETAILED DESCRIPTION

[0022] In order to enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings 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 making creative efforts should fall within the scope of protection of the present invention.

[0023] As described in the background technology, existing big data-based traffic analysis cloud platforms have problems such as complex data processing, highly subjective weighting of evaluation indicators, and inability to adjust according to real-time dynamics.

[0024] In response to the above technical issues, such as Figure 1 As shown, the present invention introduces a traffic analysis cloud platform based on big data, including a basic traffic data collection / access module, a traffic data management module, a traffic data processing module, a big data analysis module, a visualization and business application module, a data security management module and a platform basic service module, wherein; the basic traffic data collection / access module, the basic traffic data collection / access module is used to collect multi-source traffic data and classify and store it; the traffic data management module is connected to the basic traffic data collection / access module, the traffic data management module is used to set the data specifications and storage location for collecting multi-source traffic; the traffic data processing module is connected to the traffic data management module, the traffic data processing module is used to pre-process the data; the big data analysis module is connected to the traffic data processing module, the big data analysis module is used to perform basic analysis on the pre-processed data and generate analysis results; the visualization and business application module is connected to the big data analysis module, the visualization and business application module is used to display the analysis results and provide business applications; the platform basic service module is connected to the aforementioned modules, and the platform basic service module provides basic services to the entire cloud platform; the data security management module is connected to the aforementioned modules, and is used to ensure the data security of the cloud platform.

[0025] In a further embodiment, the basic traffic data collection / access module uses various technical means and devices to collect basic traffic-related information and data and uploads it to various interfaces of the cloud platform. Basic traffic data includes static data, dynamic data, and user behavior data. Static data includes road network topology, point of interest (POIs), and transportation facility locations; dynamic data includes GPS tracks, ETC gantry records, video surveillance, geomagnetic coils, and meteorological sensor data; and user behavior data includes bus / subway card swipe records, navigation app requests, and ride-sharing order data. Technical tools implementing this functionality in this embodiment may include: collecting sensor data using IoT protocols (MQTT / CoAP); receiving high-concurrency data streams in real time through distributed message queues (Kafka / Pulsar); and using web crawlers to capture open internet data (such as AutoNavi / Baidu traffic information).

[0026] In a further embodiment, the traffic data management module stores the data collected by the collection system and passes it to the traffic data processing module. The preprocessing of the traffic data processing module includes data warehouse planning, data cleaning, data filling and data integration of basic traffic data, wherein data cleaning includes data denoising to filter drift points in the data, the data filling includes using a difference algorithm to fill missing sensor values, and the data integration includes unifying data of different frequencies to the same timestamp and coordinate system. The implementation of data cleaning can be achieved through Spark Structured Streaming, and abnormal events (such as traffic accidents) can be detected through Flink CEP (complex event processing).

[0027] In a further embodiment, Figure 2 As shown in the figure, the big data analysis module includes: a basic analysis unit, a machine learning model unit, a graph computing application unit, a stream processing engine unit and an edge computing optimization unit. Among them, the basic analysis unit is used to count the traffic volume and average speed of a road section according to the time window, mine commuting patterns through subway card swiping data, and classify congestion levels based on speed thresholds or congestion indexes; the machine learning model unit includes a short-term road condition prediction model based on LSTM / Transformer, a traffic accident recognition model based on CNN to improve recognition accuracy, a ride-hailing dispatching strategy optimization model based on reinforcement learning (DQN) to reduce the empty driving rate, and a traffic policy simulation and deduction model based on SUMO+AI Agent; the graph computing application unit identifies key nodes in the road network based on Neo4j and uses GraphSAGE to simulate the congestion propagation path; the stream processing engine unit adopts Flink for stateful computing and combines Kafka for event distribution; the edge computing optimization unit deploys the YOLOv5 model on the NVIDIA Jetson edge device to realize real-time analysis of video streams and control the delay within the preset value. In this embodiment, the analysis process of the big data analysis module includes: basic analysis of the data to describe statistics, analysis of the distribution / trend of traffic data, and creation of association rules between traffic data based on the above data by using the Apriori algorithm. Based on the association rules created for traffic data, ARIMA / LSTM can be used for time series prediction of subsequent traffic conditions. In this embodiment, machine learning is achieved through integrated learning of XGBoost feature importance, deep learning through the Transformer model, and reinforcement learning of the model using a dynamic pricing strategy. User portraits can be analyzed through data modeling based on the RFM model, and knowledge graphs can be established through entity relationship mining. Causal inference can also be performed through the double difference method.

[0028] In a further embodiment, Figure 3As shown, the visualization and business application module includes a visualization tool chain unit and a business system integration unit. The visualization tool chain unit includes: a basic display component, a spatiotemporal interaction component, and a three-dimensional simulation component. The basic display component uses Tableau / Power BI to generate a traffic flow heat map. The traffic flow heat map presents the statistical results of traffic data in an intuitive two-dimensional graphic, focusing on displaying traffic distribution and trend characteristics; the spatiotemporal interaction component uses Kepler.gl to draw a vehicle trajectory flow map. The spatiotemporal interaction component analyzes the dynamic correlation of traffic data based on the spatiotemporal dimension, and mines behavioral patterns and spatial interaction characteristics; the three-dimensional simulation component uses CesiumJS to build a digital twin road network, which is used to realize three-dimensional visualization and dynamic simulation deduction of traffic scenarios. In this embodiment, the business system integration unit includes a signal control interface, an emergency command interface, and a public service interface. The signal control interface connects to the SCATS signal system through an API and issues an optimization plan; the emergency command interface pushes the accident prediction results to the traffic police command screen, wherein the response time for the prediction results to be pushed to the traffic police command screen is lower than the preset response time threshold; the public service interface publishes real-time traffic conditions through the AutoNavi API to guide users to avoid congestion.

[0029] In further embodiments, the data security management module implements functions including access control, data encryption, data backup and recovery, security auditing, data desensitization, security monitoring, and early warning. The data security management module also functions to rationally partition and isolate the cloud platform's internal network. The platform's basic service module implements basic services including at least database middleware, a TCP service framework, an authentication center, a service center, unified monitoring, a configuration center, a message queue, a task scheduling platform, a distributed cache, a file service, a logging platform, a development interface platform, a distributed deployment platform, and a development API gateway.

[0030] In a further embodiment, Figure 4 As shown, a traffic analysis method based on big data includes the following steps: S1. Collect sensor data through the Internet of Things protocol, use distributed message queues to process high-concurrency data streams, and simultaneously capture Internet traffic data; In this embodiment, the collected data includes sensor data, camera data, and comprehensive data. Sensor data can be obtained through geomagnetic sensors, radars, RFID, etc., and is used to collect information such as vehicle speed, flow, and location, and is usually transmitted through MQTT or CoAP protocols. Camera data: Real-time images of traffic scenes are obtained through video streaming (such as RTSP protocol) or image capture technology (such as OpenCV) for license plate recognition and accident detection. Comprehensive data sources: Include GPS trajectory data, meteorological data, social media events, etc., which are uploaded to the cloud through API interfaces or log files.

[0031] In this embodiment, data upload can use Python's paho-mqtt library to implement sensor data upload: the implementation code includes Python import paho.mqtt.client as mqtt client = mqtt.Client() client.connect("cloud-server", 1883) client.publish("sensor / traffic", payload=json.dumps(sensor_data)).

[0032] S2: Store structured data in a distributed database or time series database, and unstructured data in an object storage service. Optimize data classification and retrieval through a metadata management system. In this embodiment, the data collected is stored according to the type of data it needs to be stored in: structured data (such as sensor readings) is stored in a distributed database (such as HBase, Cassandra) or a time-series database (such as InfluxDB). Unstructured data (such as videos and images) uses an object storage service (such as AWS S3, MinIO). Metadata management: Data classification and retrieval optimization are achieved through Apache Atlas or a custom metadata tag system.

[0033] Use HBase's Java API to store structured data. The implementation code includes: java Configuration config = HBaseConfiguration.create(); Connection connection = ConnectionFactory.createConnection(config); Table table = connection.getTable(TableName.valueOf("traffic_data")); Put put = new Put(Bytes.toBytes("row_key")); put.addColumn(Bytes.toBytes("cf"), Bytes.toBytes("speed"), Bytes.toBytes(60)); table.put(put); S3 uses a stream processing engine for real-time data cleaning and statistics, combined with machine learning models for road condition prediction and accident detection; In this embodiment, real-time analysis uses Apache Flink or Spark Streaming to process real-time traffic flow statistics and abnormal event (such as accident) detection.

[0034] Offline analysis uses MapReduce or Spark to mine historical data, such as congestion pattern prediction and road network optimization modeling.

[0035] The AI model application is based on TensorFlow / PyTorch to train the traffic prediction model and integrate it into the analysis pipeline.

[0036] Flink Real-time Traffic Statistics Example (Scala). The code to implement the traffic statistics example can be: Scala val env = StreamExecutionEnvironment.getExecutionEnvironment val trafficStream = env.addSource(new KafkaSource[TrafficData]()) trafficStream.keyBy(_.roadId) .window(TumblingProcessingTimeWindows.of(Time.minutes(5))) .sum("vehicleCount") .addSink(new KafkaSink[Result]()).

[0037] S4. Generate traffic situation visualization results through visualization tools and connect to business systems to achieve dynamic control; In this embodiment, the traffic management department can display real-time road conditions and accident heat maps through a dashboard (such as Grafana); and push early warning information (such as congestion and construction) through the API.

[0038] The vehicle computer / navigation system pushes dynamic route planning suggestions based on Kafka or gRPC and optimizes the navigation algorithm based on historical data.

[0039] Use FastAPI to build a RESTful interface for traffic management departments to call. The code to implement the call can be: Python from fastapi import FastAPI app = FastAPI() @app.get(" / traffic / alerts") def get_alerts(area: str): return query_alerts_from_db(area).

[0040] S5. Use encryption technology to ensure data transmission and storage security, implement fine-grained access control based on the permission control model, and cope with traffic peaks through containerized deployment and elastic expansion mechanisms.

[0041] In further embodiments, the traffic analysis platform of the present invention can be applied to urban traffic management and optimization, public transportation service and operation optimization, smart highway and road management, port logistics and autonomous driving, cross-regional collaboration and emergency management, traffic planning and decision support, as well as traffic law enforcement and credit supervision. Figures 5 and 6 The figure shows a schematic diagram of the structure of the present invention when it is applied. When it is applied, the layout of smart sensors is as follows: a variety of smart sensors are densely deployed at key nodes of urban roads, such as intersections, tunnel entrances, and viaducts. For example, geomagnetic sensors can accurately detect data such as vehicle passing time, speed, and flow, providing basic information for traffic flow analysis. Taking some sections of Chang'an Avenue in Beijing as an example, through the reasonable layout of geomagnetic sensors, it is possible to obtain real-time traffic data of thousands of vehicles per hour, providing strong support for traffic management departments to accurately grasp the traffic flow conditions on the road, thereby realizing dynamic adjustment of traffic light duration according to actual traffic flow, reducing vehicle waiting time, and improving road traffic efficiency.

[0042] The deployment of high-definition surveillance cameras includes installing them along major urban roads and at transportation hubs. These cameras not only monitor traffic conditions in real time but also enable license plate recognition and vehicle classification. For example, the Shanghai Hongqiao Transportation Hub, through its large number of high-definition cameras combined with intelligent image recognition technology, can accurately identify various vehicles within a short period of time, automatically capturing and issuing warnings for illegal parking and lane changes. This also provides more intuitive data support for traffic flow analysis, helping to promptly identify and address traffic congestion points.

[0043] On-board smart terminals: Taxis, buses, logistics vehicles, and other operating vehicles are equipped with on-board smart terminals, which use GPS positioning systems to upload real-time information such as vehicle location, speed, and trajectory. For example, after Guangzhou buses are equipped with on-board smart terminals, bus companies can monitor the operating status of each bus in real time, flexibly adjust departure times and routes based on road conditions and passenger flow, and improve bus operation efficiency and service quality. Passengers can also use mobile apps to view the bus's real-time location and estimated arrival time, allowing them to plan their travel schedules accordingly.

[0044] Big Data Analysis Module: This module leverages a big data analysis system to conduct in-depth mining and analysis of massive amounts of collected traffic data. For example, Didi Chuxing, by analyzing the travel data of hundreds of millions of users on its platform, can accurately understand the distribution patterns of travel demand in different urban areas and at different time periods. This allows it to provide intelligent dispatching services for taxi and ride-hailing drivers, improving order acceptance efficiency and reducing idle mileage. It also provides data support to urban transportation planners, helping them optimize urban transportation network layouts.

[0045] Traffic models: Traffic prediction models are built based on machine learning and deep learning algorithms to accurately predict traffic flow, congestion, and other conditions. For example, Baidu Maps uses its traffic prediction model, combined with historical data and real-time traffic information, to predict the probability and degree of congestion on major urban roads in the next few hours. This provides users with intelligent navigation services, guides users to avoid congested roads, and plans the best travel routes, effectively alleviating urban traffic pressure. Traffic command: Intelligent traffic command systems that integrate multiple software functions can achieve intelligent and coordinated traffic management. For example, at Shenzhen's traffic command center, the intelligent traffic command system can integrate multiple information sources such as traffic flow data, surveillance video information, and traffic accident alarms in real time. Through comprehensive analysis and processing using intelligent algorithms, it automatically generates traffic command and dispatch plans, such as timely adjusting signal light timing and directing traffic police to congestion or accident scenes for diversion, thus achieving efficient management and precise regulation of urban traffic.

[0046] like Figure 7The following data shows application data from this invention in various cities, demonstrating significant optimization and improvement in multiple traffic aspects compared to pre-implementation scenarios. This invention offers the following advantages over existing technologies: real-time perception and intelligent decision-making. Holographic data acquisition includes real-time acquisition of traffic flow, speed, and vehicle characteristics through multiple sources, including cameras, geomagnetic coils, and GPS. AI-driven decision-making utilizes deep learning models to predict traffic patterns and optimize autonomous vehicle scheduling, resulting in a 30% increase in overall efficiency.

[0047] Refined analysis and precision services include vehicle feature recognition: supporting over 2,000 detailed features such as license plates, vehicle models, and vehicle logos, and enabling image-based search, helping police quickly locate vehicles with cloned license plates and hit-and-run vehicles. Travel service optimization: Through cross-departmental data integration, illegal vehicles are targeted. The public transportation senior citizen discount card management system effectively reduces card fraud through data comparison.

[0048] Cross-system collaboration and emergency response, including multi-departmental coordination: A "three-screen linkage" mechanism integrates traffic police, transportation, and market supervision data to achieve a closed-loop process for illegal vehicle operations, from early warning to enforcement. Rapid emergency response: The cloud control platform can detect an incident within 30 seconds and initiate a multi-party coordinated response.

[0049] Cost optimization and sustainable development are achieved through hardware reuse and automated operations. This technology supports the reuse of older servers, reducing initial investment, while automated operations tools reduce labor costs. Self-driving electric container trucks and an intelligent dispatching system improve single-axle operating efficiency while reducing energy consumption.

[0050] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0051] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1a process or multiple processes and / or boxes Figure 1 A system that specifies the functions of a box or boxes.

[0052] These computer program instructions may also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture including an instruction system that is implemented in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0053] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0054] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above and that the invention can be embodied in other specific forms without departing from the spirit or essential characteristics of the invention. Therefore, the embodiments should be considered in all respects as illustrative and non-restrictive, and the scope of the invention is defined by the appended claims, not the foregoing description, and all variations within the meaning and range of equivalents of the claims are intended to be included therein. Any reference sign in a claim should not be construed as limiting the claim to which it relates.

[0055] In addition, it should be understood that although this specification is described in terms of implementation methods, not every implementation method contains only one independent technical solution. This narrative method of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment can also be appropriately combined to form other implementation methods that can be understood by those skilled in the art.

Claims

1. A traffic analysis cloud platform based on big data, characterized by: The traffic analysis cloud platform includes: basic traffic data collection / access module, traffic data management module, traffic data processing module, big data analysis module and visualization and business application module; The basic traffic data collection / access module is used to collect, classify and store multi-source traffic data; The traffic data management module is connected to the basic traffic data collection / access module, and the traffic data management module is used to set the data specifications and storage location for collecting multi-source traffic; The traffic data processing module is connected to the traffic data management module, and the traffic data processing module is used to pre-process the data; The big data analysis module is connected to the traffic data processing module, and the big data analysis module is used to perform basic analysis on the pre-processed data and generate analysis results; The visualization and business application module is connected to the big data analysis module, and the visualization and business application module is used to display analysis results and provide business applications.

2. A traffic analysis cloud platform based on big data according to claim 1, characterized in that: The basic traffic data collection / access module is used to collect basic information data related to traffic and upload it to various interfaces of the cloud platform. The basic traffic data includes static data, dynamic data and user behavior data, among which, The static data includes road network topology, POI and transportation facility locations; The dynamic data includes GPS tracks, ETC gantry records, video surveillance, geomagnetic coils and meteorological sensor records; The user behavior data includes bus / subway card swiping records, navigation APP requests, and shared travel order data.

3. A traffic analysis cloud platform based on big data according to claim 1, characterized in that: The preprocessing of the traffic data processing module includes data warehouse planning, data cleaning, data filling and data integration of basic traffic data. Data cleaning includes data denoising to filter drift points in the data. The data filling includes using the difference algorithm to fill the missing sensor values. The data integration includes unifying data of different frequencies to the same timestamp and coordinate system.

4. The traffic analysis cloud platform based on big data according to claim 1 is characterized in that: The big data analysis module includes: a basic analysis unit, a machine learning model unit, a graph computing application unit, a stream processing engine unit, and an edge computing optimization unit, wherein: The basic analysis unit is used to count the traffic volume and average speed of road sections by time window, mine commuting patterns through subway card swipe data, and classify congestion levels based on speed thresholds or congestion indexes; Machine learning model units, including a short-term traffic condition prediction model based on LSTM / Transformer, a traffic accident identification model based on CNN, a ride-hailing dispatch strategy optimization model based on reinforcement learning, and a traffic policy simulation model based on SUMO+AI Agent; The graph computing application unit uses Neo4j to identify key nodes in the road network and uses GraphSAGE to simulate congestion propagation paths; The stream processing engine unit uses Flink for stateful computing and combines it with Kafka for event distribution. The edge computing optimization unit deploys the YOLOv5 model on the NVIDIA Jetson edge device to achieve real-time analysis of video streams and control the latency within the preset value.

5. The traffic analysis cloud platform based on big data according to claim 1 is characterized in that: The visualization and business application module includes a visualization tool chain unit and a business system integration unit, wherein the visualization tool chain unit includes: a basic display component, a spatiotemporal interaction component and a three-dimensional simulation component, wherein, The basic display component uses Tableau / Power BI to generate a traffic flow heat map, which presents the statistical results of traffic data in an intuitive two-dimensional graphic, focusing on traffic distribution and trend characteristics; The spatiotemporal interaction component uses Kepler.gl to draw a vehicle trajectory flow diagram. The spatiotemporal interaction component analyzes the dynamic correlation of traffic data based on the spatiotemporal dimension and mines behavioral patterns and spatial interaction features. The three-dimensional simulation component constructs a digital twin road network through CesiumJS, and the digital twin road network is used to realize three-dimensional visualization and dynamic simulation deduction of traffic scenes.

6. A traffic analysis cloud platform based on big data according to claim 5, characterized in that: The business system integration unit includes a signal control interface, an emergency command interface and a public service interface, wherein: The signal control interface connects to the SCATS signal system through the API and issues optimization solutions; The emergency command interface pushes the accident prediction results to the traffic police command screen, wherein the response time of pushing the prediction results to the traffic police command screen is lower than a preset response time threshold; The public service interface publishes real-time traffic conditions through the AutoNavi API to guide users to avoid congestion.

7. The traffic analysis cloud platform based on big data according to claim 1 is characterized in that: It also includes a data security management module, which is connected to the aforementioned modules and is used to ensure the data security of the cloud platform. The functions of the data security management module include access control, data encryption, data backup and recovery, security auditing, data desensitization, security monitoring and early warning; the data security management module is also used to reasonably divide and isolate the internal network of the cloud platform.

8. The traffic analysis cloud platform based on big data according to claim 1 is characterized in that: It also includes a platform basic service module, which is connected to the aforementioned modules. The platform basic service module provides basic services for the entire cloud platform. The basic services include at least database middleware, TCP service framework, authentication center, service center, unified monitoring, configuration center, message queue, task scheduling platform, distributed cache, file service, log platform, development interface platform, distributed deployment platform and development API gateway.

9. A traffic analysis method based on big data, the traffic analysis method being applied to the traffic analysis cloud platform based on big data according to any one of claims 1 to 8, characterized in that: The traffic analysis method comprises the following steps: Collect sensor data through IoT protocols, use distributed message queues to process high-concurrency data streams, and simultaneously capture Internet traffic data; Store structured data in a distributed database or time series database, and unstructured data in an object storage service. Use a metadata management system to optimize data classification and retrieval. Use stream processing engines for real-time data cleaning and statistics, combined with machine learning models for road condition prediction and accident detection; Generate traffic situation visualization results through visualization tools and connect to business systems to achieve dynamic control; Encryption technology is used to ensure data transmission and storage security, fine-grained access control is implemented based on the permission control model, and traffic peaks are handled through containerized deployment and elastic expansion mechanisms.

10. A computer-readable storage medium, characterized in that The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the traffic analysis method based on big data according to claim 9 or executes the traffic analysis cloud platform based on big data according to any one of claims 1-8.

Images