Multi-terminal collaborative urban traffic monitoring and management system

Abstract

The invention relates to the field of traffic management, in particular to a multi-terminal cooperative urban traffic monitoring and management system. Comprising a deep learning end, a client and an embedded end, the client comprises a front end and a rear end; the embedded terminal is used for receiving data of the sensor and controlling traffic signal facilities; the deep learning end comprises a pedestrian and traffic flow recognition module and a pedestrian and traffic flow prediction module; the pedestrian and traffic flow recognition module is used for recognizing and obtaining traffic flow real-time data by using a deep learning model according to data of a traffic camera in the server, and storing the traffic flow real-time data to the server; the pedestrian and traffic flow prediction module trains a prediction model according to the traffic flow data, and obtains traffic flow prediction data through the prediction model. According to the invention, in a changeable and complex traffic environment, flow data can be precisely processed, signal lamps can be dynamically regulated and controlled, and emergencies can be quickly responded.

Description

Technical Field

[0001] The present invention relates to the field of traffic management, and particularly to a multi-terminal collaborative urban traffic monitoring and management system. Background Art

[0002] In the field of intelligent traffic management, traffic flow recognition and prediction technologies based on the collaborative work of deep learning, embedded platforms, and client systems have become important tools for improving road management efficiency and optimizing traffic flow. With the increasing requirements for the complexity and real-time nature of traffic management, although there have been some advancements in existing technologies, they perform poorly in the face of high-density traffic flows and emergencies in practical applications, and still face a series of challenges in system integration.

[0003] Domestic research, such as the traffic flow monitoring system based on infrared sensors proposed by An Leyao et al., can effectively monitor traffic flow in certain environments. However, due to the limitation of infrared accuracy, there are large errors in the detection of high-density traffic flows and low-speed or stationary vehicles, affecting its accuracy under complex road conditions. Poor performance in complex traffic environments leads to a significant reduction in the accuracy of traffic flow recognition and prediction, thereby affecting the regulation effect of traffic lights. In addition, the system lacks a complete client platform and fails to provide traffic managers with real-time and intuitive traffic flow data analysis and visualization, further reducing the effectiveness and timeliness of the decision-making support system.

[0004] The traffic flow prediction system based on an embedded platform proposed by Wang Ming et al. predicts traffic flow through machine learning methods and uses embedded devices for data collection and preliminary processing. Although this system can perform a certain degree of traffic flow prediction, due to the lack of effective integration of the client platform, traffic managers cannot obtain real-time traffic flow data and prediction information that changes dynamically, which seriously affects the timeliness and accuracy of decision-making in the face of complex or sudden traffic conditions.

[0005] Foreign research, such as the traffic signal control system based on embedded intelligent hardware proposed by Li et al., can theoretically regulate traffic flow efficiently through an embedded platform. However, due to the high hardware cost and large integration difficulty of the system, the complexity of actual deployment and maintenance makes it difficult to popularize in large-scale urban traffic management. At the same time, the system lacks effective integration with the client system and cannot provide traffic managers with real-time traffic flow data monitoring and analysis, further limiting its practical application in complex traffic environments.

[0006] Similarly, the adaptive traffic signal control system proposed by Mitra et al. can dynamically adjust signal timing according to traffic flow to optimize traffic flow. However, the lack of deep integration with the embedded platform and the client has led to a slow response of the system when dealing with traffic flow fluctuations or sudden traffic events, and the response stability of the system is insufficient. The drawback of this technology is the lack of a perfect real-time data feedback and an interactive interface, making it difficult for managers to make timely decisions, resulting in insignificant optimization of traffic flow.

[0007] In summary, there are many deficiencies in the prior art in the collaborative work of deep learning, embedded platforms, and clients. The systems generally face problems such as high hardware costs, complex system integration, and insufficient real-time response, and it is difficult to effectively meet the increasingly complex urban traffic management needs. Especially in the aspects of traffic flow monitoring, prediction, and signal regulation, the lack of an efficient combination of deep learning and embedded platforms limits the accuracy and real-time performance of traffic flow prediction. Summary of the Invention

[0008] In view of the above problems, the present invention aims to solve the problems of computational complexity, hardware cost, and system integration in the prior art through the collaborative work of deep learning, embedded platforms, and clients, improve the accuracy and real-time response ability of traffic flow recognition and prediction, and promote the intelligent traffic management system to develop in a more efficient, stable, and intelligent direction.

[0009] A multi-terminal collaborative urban traffic monitoring and management system, comprising: a deep learning terminal, a client, and an embedded terminal;

[0010] Among them, the client includes a front end and a back end. The front end is used to display specified data to traffic participants according to the settings of the administrator and for the administrator to manage the system. The management includes setting rules and querying data. The back end includes a server, and the server includes a database for storing data;

[0011] The embedded terminal is used to receive data from sensors and control traffic signal facilities; the sensors include traffic cameras; the embedded terminal stores the data received from the sensors in the server;

[0012] The deep learning terminal includes a vehicle and pedestrian flow recognition module and a vehicle and pedestrian flow prediction module; the vehicle and pedestrian flow recognition module is used to identify and obtain real-time traffic flow data using a deep learning model based on the data of traffic cameras in the server, and store the real-time traffic flow data in the server; the vehicle and pedestrian flow prediction module trains a prediction model based on the saved real-time traffic flow data, and obtains traffic flow prediction data through the prediction model. The traffic flow prediction data is saved in the server;

[0013] The embedded end controls the traffic signal facilities according to the rules set by the administrator through the front end, as well as traffic flow prediction data, real-time traffic flow data, and sensor data.

[0014] Preferably, the front end includes a road traffic guidance screen and a web page end. The road traffic guidance screen is used to display data to the public, and the web page end is for the administrator to manage the system.

[0015] Preferably, the traffic signal facilities include an LED-buzzer module. Among the real-time traffic flow data, vehicle speed information is included.

[0016] The embedded end controls the LED-buzzer module to send light and sound reminders to traffic participants according to the rules set by the administrator through the front end and the vehicle speed information.

[0017] Preferably, the sensor includes a light intensity sensor; the traffic signal facilities include a traffic light module. The embedded end adjusts the brightness of the traffic light module according to the rules set by the administrator through the front end and the light data of the light intensity sensor.

[0018] Preferably, the sensor includes a meteorological environment monitoring module; the front end displays the data of the meteorological environment monitoring module to traffic participants according to the administrator's settings.

[0019] Preferably, the sensor includes a rain drop sensor. The front end reminds traffic participants of the risk of slippery roads according to the administrator's settings and the data of the rain drop sensor.

[0020] Preferably, the traffic facilities include a traffic light module; the embedded end adjusts the timing of the traffic light module according to the rules set by the administrator and the traffic flow prediction data, real-time traffic flow data.

[0021] Beneficial effects: The present invention proposes an intelligent traffic flow recognition and signal light optimization control system that collaborates based on deep learning, an embedded platform, and a client system, with particular emphasis on the efficient collaboration between the embedded platform and the client system. By optimizing the interaction between the hardware platform and the client, the system significantly reduces the dependence on high-performance hardware, improves the real-time management efficiency of the system, enabling it to accurately process traffic data, dynamically control signal lights, and quickly respond to emergencies in a changing and complex traffic environment.

[0022] In the present invention, deep learning algorithms are used to accurately identify and predict traffic flow, and the identification and prediction results are stored in a database in real time. The client system can read the latest traffic flow information, signal light status, and prediction results by accessing the real-time data in the database, thereby realizing data interconnection and real-time interaction. This mechanism not only ensures the consistency and accuracy of data, but also greatly improves the flexibility, real-time response ability, and overall operation efficiency of the system.

[0023] In addition, the present invention introduces two types of intelligent client systems - the road traffic guidance screen system and the Web page client system - to further enhance the interactivity and real-time performance of the system. The guidance screen system dynamically displays key data such as road traffic events, real-time traffic conditions, signal light status, congestion index, and traffic safety tips by interacting with the server in real time.

[0024] At the same time, the Web page client system provides a convenient operation interface for traffic management personnel, supporting functions such as traffic flow monitoring, prediction analysis, signal light scheduling, and emergency response. Through highly integrated data analysis functions, the Web client system can display traffic flow data in real time, optimize signal light timing strategies in combination with traffic flow prediction models, help management personnel make decisions quickly, and improve road traffic efficiency. Adopting a front-end and back-end separation architecture enables the system to process large-scale data streams and provide real-time feedback on traffic changes, effectively enhancing the intelligence and decision-making support capabilities of traffic management.

[0025] By combining the high-efficiency computing power of the embedded platform with the real-time interaction of the intelligent client system, the present invention not only effectively reduces the limitations of the prior art in terms of computational complexity, hardware cost, system integration, etc., but also significantly improves the accuracy, real-time performance, and prediction ability of traffic flow identification. The system can be widely applied in the field of urban traffic management to optimize traffic flow regulation and resource allocation, improve the intelligent level of traffic management, and promote the sustainable development of smart cities. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] Figure 1 is the overall structure diagram of the present invention;

[0027] Figure 2 is the technical solution diagram of the embedded system;

[0028] Figure 3 is the technical solution diagram of the road traffic guidance screen system;

[0029] Figure 4 is the design block diagram of the road traffic guidance screen system;

[0030] Figure 5 is the traffic signal light regulation flow chart. DETAILED DESCRIPTION OF THE INVENTION

[0031] To make the objectives, features, and advantages of the present invention more apparent and understandable, the following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the embodiments described below are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0032] 1. Overall System Technical Solution

[0033] As Figure 1 shown, this embodiment is an intelligent traffic flow monitoring and management solution based on the collaborative work of deep learning, an embedded platform, and a client system. Combining the STM32F407ZGT6 microprocessor and the Jetson Nano mainboard, an efficient traffic management framework is constructed. The system, through innovative software and hardware co-design, makes full use of the data acquisition and processing capabilities of the embedded platform to monitor and regulate traffic flow in real time.

[0034] The Jetson Nano obtains the video stream from the USB camera through MJPG-Streamer and transmits it to the database server in real time. The system accurately detects and continuously tracks the vehicle and pedestrian targets in the video through a deep learning model, realizes the real-time monitoring of traffic flow, and stores the detection and tracking results in the database in real time. At the same time, the system introduces a time series prediction model to accurately capture the traffic flow change trend, provides forward-looking traffic flow prediction for the adaptive regulation of traffic lights, and significantly improves the accuracy and intelligence level of traffic management.

[0035] The hardware part of the system relies on the STM32F407ZGT6 microprocessor and is equipped with a variety of sensors (such as a rain drop sensor, a light intensity sensor, etc.) to realize the real-time data acquisition of the traffic environment. The data collected by these sensors is transmitted to the Jetson Nano through serial communication and uploaded to the database of the server through the HTTP protocol to ensure the efficiency and reliability of data transmission.

[0036] In terms of the client system, the Web client, through a front-end and back-end separation architecture, supports users to query traffic flow data in real time and visually displays information such as traffic flow, signal timing, and traffic prediction to help traffic management personnel make decisions. The Web client not only supports the display of traffic flow data, but also, through an integrated control panel, allows users to issue control instructions through an intuitive interface and send them to the embedded end through the MQTT protocol, thereby adjusting traffic signal timing and road settings to optimize traffic flow and traffic efficiency.

[0037] In addition, the system is also equipped with a road traffic guidance screen, which obtains data from the server in real time through HTTP requests and dynamically displays information such as traffic events, road congestion, and real-time weather. This system can provide drivers with instant traffic information and offer route adjustment suggestions based on the current road conditions to ensure road safety and smooth traffic flow.

[0038] 2. Technical Solution of the Deep Learning End System

[0039] The deep learning end system of this embodiment is divided into a vehicle and pedestrian flow recognition module and a vehicle and pedestrian flow prediction module.

[0040] The vehicle and pedestrian flow recognition module obtains video data in real time, uses a deep learning model to accurately identify vehicle and pedestrian targets in the video, and combines a tracking algorithm to continuously track the detection results, comprehensively monitoring traffic flow. At the same time, relevant recognition and tracking data are stored in the database in real time for client display and data analysis and processing.

[0041] The vehicle and pedestrian flow prediction module further integrates a time series prediction model based on deep learning. Through the analysis of the recognition record data, it provides long-term perspective flow prediction for the adaptive control of traffic lights, helping traffic managers make more scientific and efficient decisions. All prediction data are also stored in real time for effective management and ready access.

[0042] Through this series of intelligent data processing and analysis means, the present invention significantly improves the accuracy and intelligence level of traffic management decisions, promotes the in-depth application of traffic flow analysis and prediction, and enhances the real-time response ability and optimization efficiency of the traffic system.

[0043] 3. Technical Solution of the Embedded End System

[0044] Such as Figure 2As shown in the figure, in this embodiment, the embedded system main body adopts the STM32F407ZGT6 microcontroller and the NVIDIA Jetson Nano main board, which drive a variety of sensors to collect humidity, light intensity and meteorological environment data of the traffic road. Through the digital tube and LED lamp module, the simulation display of traffic lights is realized. In addition, the system uses the LED-buzzer module to send warning prompts to crossing pedestrians when speeding vehicles are identified, so as to improve pedestrian safety. The STM32F407ZGT6 and Jetson Nano use the serial communication protocol for data transmission, ensuring the efficient flow of information. As an edge computing platform, Jetson Nano undertakes the processing and transmission of complex video images and related data, and integrates a video acquisition module, a WiFi communication module and a power supply module to ensure real-time communication and collaboration among system modules. Through this architecture design, the system realizes the real-time collection, transmission and analysis of traffic data, provides efficient technical support for intelligent traffic management, and significantly improves the intelligent and automated level of traffic management.

[0045] a. Serial communication function module

[0046] The serial communication module realizes bidirectional data transmission between the STM32 and Jetson Nano through the RX (receive) and TX (transmit) pins of the STM32F407ZGT6 microcontroller. In this communication architecture, the TX pin of the STM32F407ZGT6 is responsible for sending data to the RX pin of the Jetson Nano, and vice versa, and data is transmitted from the TX pin of the Jetson Nano to the RX pin of the STM32F407ZGT6. This module is based on the asynchronous serial communication protocol and uses standardized baud rate, data bits, stop bits and parity bit settings to ensure the stability and accuracy of the data transmission process.

[0047] b. Traffic signal display module

[0048] The traffic signal display module consists of a two-digit digital tube, a red and green light module, an LED color light module, a light intensity sensor and a relay. The real-time countdown time is displayed through the two-digit digital tube, the red and green light module provides an indication of the traffic signal status, and the LED color light module is responsible for the dynamic adjustment of the signal light brightness. By receiving instructions and setting parameters from the web page, the countdown display and the red and green light switching logic are updated in real time to ensure the normal operation of the traffic signal. The light intensity sensor is used to detect the light intensity of the external environment and upload the data to the database. At the same time, the system realizes brightness adjustment based on the working mode and threshold set by the web page. The relay, as a control unit, dynamically adjusts the output brightness of the LED color light module.

[0049] c. Meteorological environment monitoring module

[0050] The meteorological environment monitoring module uses the M702 all-in-one sensor. The sensor transmits data to the STM32F407ZGT6 microcontroller through the serial port, which parses and formats the data. The processed data is sent to the Jetson Nano platform through the serial port for further processing and upload, and the meteorological information is transmitted to the database in real time.

[0051] d. Road humidity monitoring module

[0052] The road humidity monitoring module uses a raindrop sensor to collect real-time road surface humidity data. The sensor senses moisture changes and outputs an electrical signal proportional to the humidity. The signal is transmitted to Jetson Nano through the serial port interface for analysis and processing, and converted into usable humidity information. The processed data is uploaded to the database.

[0053] e.Road video surveillance module

[0054] The road video monitoring module uses a driver-free USB external camera, which is connected to Jetson Nano through a USB interface. Jetson Nano runs the MJPG-Streamer streaming server, which is responsible for transmitting the camera's video stream in MJPEG format to the specified port. By mapping these two ports to the public network, the remote server can receive and monitor the visual information of the lanes and sidewalks in real time, thereby achieving effective monitoring and management of traffic conditions and pedestrian safety at crosswalks.

[0055] f. Vehicle speeding alarm module

[0056] The vehicle speeding alarm function module uses an LED-buzzer module to remind pedestrians on the crosswalk to pay attention to speeding vehicles, improving the safety of pedestrians when crossing the road. When receiving the speeding alarm command, the LED light turns on and the buzzer sounds, promptly alerting pedestrians to the potential risk of speeding.

[0057] 4. Client system technical solution

[0058] The client system of the present invention is divided into a road traffic guidance screen system and a Web page client system.

[0059] (1) Road traffic guidance screen system

[0060] The road traffic guidance screen system is based on the HTTP protocol, obtains and dynamically displays traffic data in real time, and provides key technical support for the traffic guidance system. Through an efficient data processing mechanism, the system can promptly present important information such as road names, traffic events, congestion indices, real-time time, weather forecasts, and traffic safety tips. The road name information accurately identifies the current and surrounding road sections and optimizes the driving route in combination with direction guidance; the traffic event module covers traffic accidents, road construction, and temporary control situations, and promptly issues warning messages; the congestion index intuitively reflects the traffic flow conditions of each road section through color coding, providing a reference for driving decisions; the real-time time function not only synchronizes the current moment but also provides real-time reminders in specific situations such as traffic restrictions or traffic control; the weather forecast shows information such as the current weather conditions, air quality index, maximum and minimum temperatures, wind direction, etc.; the traffic safety tips combine real-time road conditions and environmental changes to issue warnings for bad weather, speed limit reminders, and precautions for special road sections, thereby enhancing driving safety.

[0061] The system adopts a modular design based on the QT framework, demonstrating excellent scalability and flexibility. It can handle complex traffic management requirements and lay a solid technical foundation for subsequent function optimization and expansion. Through the coordinated action of these comprehensive functions, drivers can obtain accurate and real-time traffic information, thereby avoiding traffic congestion and safety hazards and significantly improving travel efficiency and safety. The technical solution and design block diagram of the road traffic guidance screen system are as Figure 3 and Figure 4 shown.

[0062] (2)Web page client system

[0063] The Web page client adopts a front-end and back-end separation development mode. The front end is developed based on the Vue3 framework and integrates the ElementUI component library and the Echarts chart library to achieve the efficient construction of interface components and data visualization display. The back end is developed based on the Spring Boot framework, and Maven is used for project dependency management and build optimization. The system functions cover key modules such as user login, vehicle and pedestrian flow analysis, traffic signal control, road environment monitoring, and user and department management, realizing a comprehensive web page client with a three-level management structure. This design not only ensures the modularity and scalability of the system but also significantly improves development efficiency and system interactivity, meeting the requirements of complex traffic management systems.

[0064] a. Vehicle and pedestrian flow analysis module

[0065] Vehicle and pedestrian flow statistical record module

[0066] The vehicle and pedestrian flow statistics recording module reads historical and real-time data from the database and dynamically displays the changes in vehicle and pedestrian flow in the form of line charts and tables. Users can choose to view real-time flow data or aggregated data with a time interval of five minutes. At the same time, it supports efficient data import and export functions, facilitating data storage, analysis, and sharing.

[0067] Vehicle and Pedestrian Flow Data Prediction Module

[0068] The vehicle and pedestrian flow data prediction module reads the flow data predicted by the deep learning model from the database and dynamically displays the change trend of vehicle and pedestrian flow in the form of line charts and tables. Users can click the "Flow Prediction" button to trigger the re-prediction of the current flow data and update the database. At the same time, it supports efficient data import and export functions, facilitating data management and analysis.

[0069] Vehicle Type Data Display Module

[0070] The vehicle type data display module reads the statistical data of vehicle types from the database and dynamically presents the statistical quantities and proportions of various vehicle types in various forms such as pie charts, line charts, and bar charts, intuitively reflecting the data distribution characteristics and change trends.

[0071] b. Traffic Signal Regulation Module

[0072] Traffic Signal Adaptive Control Module

[0073] The traffic signal adaptive control module reads the flow data and user-set parameters from the database, combines the traffic signal timing calculation formula to dynamically generate the signal timing time, and intuitively displays it in the form of line charts and tables. At the same time, the relevant data is written into the database for subsequent calls, supporting data import and export functions.

[0074] Traffic Signal Manual Regulation Module

[0075] The interface of the traffic signal manual regulation module displays the current system time in real-time and synchronously shows the current green light timing duration of the sidewalk and the lane. It supports users to customize the working mode of the traffic signal (automatic mode and manual mode) and the green light timing duration parameters of the sidewalk and the lane. The regulation instructions set by the user are sent to the embedded device in real-time through the MQTT protocol to achieve the instant update of the traffic signal status. At the same time, the relevant parameter data is synchronously written into the database for recording and subsequent analysis.

[0076] Traffic Signal Traffic Control Module

[0077] The traffic signal traffic control module allows users to set different traffic control states according to their needs, including all-red in all four directions, all-green in the east-west direction (all-red in the north-south direction), all-green in the north-south direction (all-red in the east-west direction), yellow flashing, and all-off in all four directions (shutdown), etc. The control commands set by the user through the interface are sent to the embedded device in real time through the MQTT protocol, so as to switch the traffic signal to the corresponding display state. At the same time, the set mode data is synchronously written into the database.

[0078] Traffic signal brightness control module

[0079] The traffic signal brightness control module allows users to customize the brightness control mode (automatic mode and manual mode) of the traffic signal and its threshold parameters. The control commands configured by the user through the interface are sent to the embedded device in real time through the MQTT protocol to achieve instant adjustment of the traffic signal brightness. At the same time, the relevant parameter data is synchronously written into the database.

[0080] c. Road environment monitoring module

[0081] Traffic road map module

[0082] The traffic road map module realizes real-time display of the road status by calling the Gaode Map API interface, and supports custom addition of component modules required for development according to needs.

[0083] Road video monitoring function module

[0084] The road video monitoring module supports users to select sidewalk and lane monitoring at different points. By obtaining the video stream data at the embedded end, the monitoring screen is displayed in real time on the web interface, ensuring that users can efficiently monitor the real-time conditions of all important positions on the road, enhancing the real-time and accuracy of traffic management.

[0085] Meteorological environment monitoring module

[0086] The meteorological environment monitoring module extracts meteorological environment data from the database, dynamically displays the environmental changes, and presents the relevant information in the form of line charts and tables. The latest measurement data is displayed on the page, and the import and export functions of the data are supported for users to conduct further analysis and processing.

[0087] Road humidity monitoring module

[0088] The road humidity monitoring module extracts road humidity data from the database, dynamically displays the humidity change trend in the form of a line chart, and at the same time supports users to customize the humidity warning threshold, which will be automatically saved to the database. The system can respond to the risk of road slipperiness in time during monitoring and give corresponding road slipperiness reminders on the road traffic guidance screen.

[0089] Road event control module

[0090] The road event regulation module allows users to configure road event types according to actual needs, including traffic accidents, road construction, road control, and no event. Relevant regulation data is synchronized to the database in real time to support dynamic warnings in systems such as road traffic guidance displays.

[0091] Weather forecast query module

[0092] The weather forecast query module is implemented by integrating the AMap.Weather weather query service plugin, which can dynamically display weather information on the page, covering the real-time weather function to obtain the current weather conditions and weather forecast data.

[0093] d. User and department management module

[0094] The user and department management function module includes the information management of individuals, employees, and departments. It adopts a three-level permission management architecture. Low-level permission users can only view traffic management and user information. Intermediate-level permission users have additional traffic management functions. High-level permission users have further added user management functions. Through hierarchical authorization, the system can finely configure the management permissions of each department, assign corresponding management functions to different roles, and thus achieve efficient collaboration and control of traffic management.

[0095] 5. System collaborative working technical solution

[0096] a. Collaboration of traffic signal regulation functions

[0097] As Figure 5 shown, the working modes of traffic signals are three modes: adaptive control, manual regulation, and traffic control. Users can select the corresponding mode through the slider component in the Web page client. At the same time, the signal brightness regulation allows users to customize the brightness regulation mode (automatic mode and manual mode) and its threshold parameters in the Web page client. The regulation instructions set by the user are sent to the embedded device in real time through the MQTT protocol to achieve instant updates of the traffic signal display and brightness status. At the same time, the relevant parameter data is synchronized to the database for recording and subsequent analysis.

[0098] b. Collaboration of signal adaptive control functions

[0099] The pedestrian and vehicle flow recognition system obtains the real-time video monitoring screen of the embedded camera, uses a deep learning model to perform high-precision recognition of pedestrians and vehicle targets in the video, and combines the tracking algorithm to achieve continuous tracking of the detection results, comprehensively monitor traffic flow and road congestion, support vehicle type recognition and classification statistics, and store relevant recognition, tracking and classification data in real time in the database. In addition, when the system detects that the speed of the approaching vehicle exceeds the set threshold, the pedestrian and vehicle flow recognition system sends a speeding alarm command in real time through the MQTT protocol. After the embedded end receives the command, the LED light turns on and the buzzer sounds, promptly alerting pedestrians to the potential risk of speeding.

[0100] Based on historical traffic data and prediction algorithm models, traffic flow within a certain time range in the future can be predicted, and relevant data will be stored in the database in real time. In addition, based on traffic flow prediction data and combined with relevant parameters configured by the user, the system can calculate the signal timing data for each time period, realize intelligent adjustment of the signal light timing of each phase, reasonably regulate the traffic flow dynamics in different directions, reduce traffic congestion, and improve traffic efficiency. The data will be synchronized to each terminal device, thereby achieving more efficient and flexible traffic management.

[0101] At the same time, the client displays traffic statistics, vehicle types, traffic forecasts and other data by reading the database.

[0102] c. Environmental data monitoring function collaboration

[0103] The embedded end collects meteorological environment and road surface humidity data in real time through the M702 all-in-one sensor and raindrop sensor. The sensor transmits the data to the STM32F407ZGT6 microcontroller through the serial port and electrical signals. The latter parses and formats the data and sends it to the Jetson Nano platform through the serial port. Jetson Nano is responsible for further processing of the data and real-time transmission to the database.

[0104] The web client extracts relevant data from the database and dynamically displays the changing trends of meteorological environment and road surface humidity in the form of line graphs and tables. It supports users to customize humidity warning thresholds and displays warning information such as slippery roads in real time on the road traffic guidance screen. The humidity thresholds set by users will be automatically saved in the database to ensure that the system can respond to slippery road risks in a timely manner during monitoring.

[0105] d. Road incident control function coordination

[0106] Users configure road event types in the Web page client according to actual needs, including traffic accidents, road construction, road control, and no events. The regulation data set by users on the interface will be synchronized to the database in real time, and the road traffic guidance screen reads the relevant data and gives dynamic warnings to ensure the immediate update and storage of control information.

[0107] The above are only the preferred embodiments of the present invention. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements should also be regarded as the protection scope of the present invention.

Claims

1. A multi-terminal collaborative urban traffic monitoring and management system, characterized in that Including: Deep learning terminal, client, and embedded terminal; Among them, the client includes a front end and a back end. The front end is used to display specified data to traffic participants according to the settings of the administrator and to enable the administrator to manage the system. The management includes setting rules and querying data. The back end includes a server, and the server includes a database for storing data; The embedded terminal is used to receive data from sensors and control traffic signal facilities; the sensors include traffic cameras; the embedded terminal stores the received sensor data in the server; The deep learning terminal includes a vehicle and pedestrian flow recognition module and a vehicle and pedestrian flow prediction module; the vehicle and pedestrian flow recognition module is used to use a deep learning model to identify and obtain real-time traffic flow data based on the data of traffic cameras in the server and store the real-time traffic flow data in the server; the vehicle and pedestrian flow prediction module trains a prediction model based on the saved real-time traffic flow data and obtains traffic flow prediction data through the prediction model, and the traffic flow prediction data is saved in the server; The embedded terminal controls the traffic signal facilities correspondingly according to the rules set by the administrator through the front end, as well as the traffic flow prediction data, real-time traffic flow data, and sensor data.

2. The multi-terminal collaborative urban traffic monitoring and management system according to claim 1, characterized in that The front end includes a road traffic guidance screen and a web page end. The road traffic guidance screen is used to display data to the public, and the web page end enables the administrator to manage the system.

3. A multi-terminal collaborative urban traffic monitoring and management system according to claim 1, characterized in that, The traffic signal facilities include an LED-buzzer module. Among the real-time traffic flow data, vehicle speed information is included; The embedded terminal controls the LED-buzzer module to issue light and sound reminders to traffic participants according to the rules set by the administrator through the front end and the vehicle speed information.

4. A multi-terminal collaborative urban traffic monitoring and management system according to claim 1, characterized in that, The sensors include light intensity sensors; the traffic signal facilities include traffic light modules. The embedded terminal adjusts the brightness of the traffic light modules accordingly according to the rules set by the administrator through the front end and the light data of the light intensity sensors.

5. A multi-terminal collaborative urban traffic monitoring and management system according to claim 1, characterized in that, The sensors include meteorological environment monitoring modules; the front end displays the data of the meteorological environment monitoring modules to traffic participants according to the settings of the administrator.

6. The multi-terminal collaborative urban traffic monitoring and management system according to claim 1, characterized in that The sensors include rain sensors. The front end reminds traffic participants of the risk of slippery roads according to the settings of the administrator and the data of the rain sensors.

7. A multi-terminal collaborative urban traffic monitoring and management system according to claim 1, characterized in that, The traffic facilities include traffic light modules; the embedded terminal regulates the signal timing of the traffic light modules according to the rules set by the administrator and the traffic flow prediction data and real-time traffic flow data.

Images