An urban road Internet of Things data management platform

Abstract

An urban road Internet of Things data management platform collects vehicle flow values ​​at multiple predetermined time points within a predetermined time period through smart street lamps deployed on predetermined roads; arranges the vehicle flow values ​​at the multiple predetermined time points into a vehicle flow time series input vector according to the time dimension; extracts time series features from the vehicle flow time series input vector to obtain vehicle flow time series features; and determines the congestion level label of the predetermined road based on the vehicle flow time series features. In this way, the congestion level of the predetermined road can be intelligently judged, providing valuable information for urban management and services, and helping to realize the monitoring, prediction and optimization of traffic congestion levels.

Description

Technical Field

[0001] The present application relates to the field of intelligent Internet of Things technology, and more specifically, to an urban road Internet of Things data management platform. Background Art

[0002] Urban road traffic congestion has always been an important issue in urban management and planning. With the acceleration of urbanization, the increase in the number of vehicles and the limited road resources have led to increasingly serious traffic congestion problems, which has brought many inconveniences to urban management and residents' travel.

[0003] However, urban traffic data is usually a large-scale, high-dimensional data set that requires effective processing and analysis to extract useful information. However, existing data management platforms can only screen and utilize data manually, and have limitations in data processing and analysis capabilities, and cannot fully utilize the potential of data. In addition, urban traffic conditions require real-time monitoring and response so that timely measures can be taken to deal with traffic congestion and incidents. However, existing data management platforms lag behind in real-time and responsiveness, and cannot provide timely traffic condition information and decision support.

[0004] Therefore, an optimized urban road IoT data management platform is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides an urban road Internet of Things data management platform, which collects the traffic flow values ​​at multiple predetermined time points within a predetermined time period by smart street lights deployed on predetermined roads; arranges the traffic flow values ​​at the multiple predetermined time points according to the time dimension as a traffic flow time series input vector; extracts the time series features of the traffic flow time series input vector to obtain the traffic flow time series features; and determines the congestion level label of the predetermined road based on the traffic flow time series features. In this way, the congestion level of the predetermined road can be intelligently judged, providing valuable information for urban management and services, and facilitating the monitoring, prediction and optimization of traffic congestion levels.

[0006] In a first aspect, a city road Internet of Things data management platform is provided, which includes:

[0007] A vehicle flow data collection module is used to collect vehicle flow values ​​at multiple predetermined time points within a predetermined time period through smart street lights deployed on predetermined roads;

[0008] A vehicle flow time series arrangement module, used for arranging the vehicle flow values ​​at the plurality of predetermined time points into a vehicle flow time series input vector according to the time dimension;

[0009] A vehicle flow time series feature analysis module, used for extracting time series features from the vehicle flow time series input vector to obtain vehicle flow time series features;

[0010] The predetermined road congestion detection module is used to determine the congestion level label of the predetermined road based on the traffic flow time series characteristics. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0012] Figure 1 This is a block diagram of an urban road Internet of Things data management platform according to an embodiment of the present application.

[0013] Figure 2A Schematic diagram of a smart street lamp according to an embodiment of the present application.

[0014] Figure 2B Schematic diagram of the application of a smart street lamp according to an embodiment of the present application.

[0015] Figure 3 This is a flowchart of a method for managing urban road Internet of Things data according to an embodiment of the present application.

[0016] Figure 4 It is a schematic diagram of the architecture of the urban road Internet of Things data management method according to an embodiment of the present application.

[0017] Figure 5 This is an application scenario diagram of the urban road Internet of Things data management platform according to an embodiment of the present application. DETAILED DESCRIPTION

[0018] The following will describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0019] Unless otherwise specified, all technical and scientific terms used in the embodiments of the present application have the same meaning as those commonly understood by those skilled in the art of the present application. The terms used in this application are only for the purpose of describing specific embodiments and are not intended to limit the scope of this application.

[0020] In the embodiments of the present application, it should be noted that, unless otherwise specified and limited, the term "connection" should be understood in a broad sense. For example, it can be an electrical connection or a connection between two components. It can be a direct connection or an indirect connection through an intermediate medium. For ordinary technicians in this field, the specific meanings of the above terms can be understood according to the specific circumstances.

[0021] It should be noted that the terms "first\second\third" involved in the embodiments of the present application are only used to distinguish similar objects, and do not represent a specific order for the objects. It is understandable that the specific order or sequence of "first\second\third" can be interchanged where permitted. It should be understood that the objects distinguished by "first\second\third" can be interchanged where appropriate, so that the embodiments of the present application described herein can be implemented in an order other than those illustrated or described herein.

[0022] After introducing the basic principles of the present application, various non-limiting embodiments of the present application will be described in detail with reference to the accompanying drawings.

[0023] Urban road traffic congestion refers to the phenomenon that vehicles travel slowly and traffic flows poorly in the urban road network due to excessive number of vehicles, limited road capacity or unreasonable traffic organization. Traffic congestion not only brings challenges to urban management and planning, but also brings many inconveniences to residents' travel.

[0024] The impact of traffic congestion on urban management and planning is mainly reflected in the following aspects:

[0025] Economic impact: Traffic congestion will cause vehicles to slow down, prolong travel time, increase transportation costs, and affect the timely delivery of goods. In addition, traffic congestion will also lead to increased fuel consumption and emissions, which will have a negative impact on the environment.

[0026] Social impact: Traffic congestion brings inconvenience to residents, increasing commuting time and stress. Long-term traffic jams may also cause traffic accidents, road conflicts and unsafe factors, affecting residents' travel safety and quality of life.

[0027] Impact on urban planning: Traffic congestion limits the development space and effectiveness of urban planning. Insufficient road capacity and poor traffic flow will limit the possibility of building new roads, improving transportation facilities and improving urban transportation networks. In addition, traffic congestion will also affect urban air quality and noise pollution, and have an adverse impact on the urban environment.

[0028] In order to solve the problem of urban road traffic congestion, it is necessary to comprehensively use means such as traffic management, urban planning, traffic facility construction and intelligent technology. Take traffic management measures, such as traffic light optimization, traffic control, traffic restrictions, etc., to regulate traffic flow and optimize traffic organization to reduce traffic congestion. Rationally plan the urban road network, increase road capacity and improve road layout to improve traffic mobility. At the same time, pay attention to public transportation construction, encourage residents to use public transportation, and reduce the use of private cars. Strengthen the construction of transportation facilities, including road reconstruction and expansion, transportation hub construction, parking facility construction, etc., to improve road capacity and traffic operation efficiency. Use intelligent transportation systems and Internet of Things technologies to monitor and analyze traffic data in real time, provide traffic status information and decision support. Through the intelligent traffic management system, it is possible to achieve intelligent optimization of traffic lights, dynamic regulation of traffic flow, etc., thereby reducing traffic congestion.

[0029] However, urban traffic data is usually a large-scale, high-dimensional data set that requires effective processing and analysis to extract useful information. Existing data management platforms are usually used for the collection, storage, processing and analysis of urban traffic data to support urban management and decision-making. The data management platform includes: Traffic sensor network platform collects traffic data in real time, such as traffic volume, vehicle speed, road conditions, etc., by deploying sensor networks on the road. The collected data is transmitted to the back-end system through wireless communication for storage and analysis. Data storage and management platform is used to store and manage large-scale traffic data. It can provide efficient data storage and retrieval functions to quickly access and process large amounts of traffic data. The data processing and analysis platform is used to process and analyze traffic data to extract useful information and insights. Various data processing and analysis algorithms, such as time series analysis, data mining, machine learning, etc., can be applied to reveal traffic conditions, trends and patterns. The visualization and decision support platform visualizes traffic data and provides interactive data query and analysis functions. It can generate maps, charts and reports on traffic conditions to help city managers and decision makers understand and evaluate traffic problems and make corresponding decisions and measures. The real-time monitoring and early warning platform can monitor traffic conditions in real time and provide real-time traffic early warning and alarm functions. It can timely detect traffic congestion, accidents and abnormal situations through real-time data updates and analysis, and send alarms and notifications to relevant departments and residents.

[0030] These data management platforms can help city managers and planners better understand and deal with traffic congestion, provide decision support and optimize urban traffic management. However, the existing platforms still have some limitations in data processing and analysis capabilities, real-time and responsiveness, and need to be further improved and perfected. However, the existing data management platforms can only screen and utilize data manually, and have limitations in data processing and analysis capabilities, and cannot fully utilize the potential of data. In addition, urban traffic conditions need to be monitored and responded to in real time so that timely measures can be taken to deal with traffic congestion and incidents. However, the existing data management platforms lag behind in real-time and responsiveness, and cannot provide timely traffic condition information and decision support.

[0031] Therefore, in this application, an optimized urban road Internet of Things data management platform is provided.

[0032] In one embodiment of the present application, Figure 1 FIG. 1 is a block diagram of an urban road IoT data management platform according to an embodiment of the present application. Figure 1 As shown, according to the urban road Internet of Things data management platform 100 of the embodiment of the present application, it includes: a vehicle flow data collection module 110, which is used to collect vehicle flow values ​​at multiple predetermined time points within a predetermined time period through smart street lights deployed on predetermined roads; a vehicle flow time series arrangement module 120, which is used to arrange the vehicle flow values ​​at the multiple predetermined time points according to the time dimension into a vehicle flow time series input vector; a vehicle flow time series feature analysis module 130, which is used to extract time series features from the vehicle flow time series input vector to obtain vehicle flow time series features; a predetermined road congestion degree detection module 140, which is used to determine the congestion degree label of the predetermined road based on the vehicle flow time series features.

[0033] In the vehicle flow data collection module 110, the location and coverage of the street lights need to be considered when deploying smart street lights to ensure that the vehicle flow data can be accurately collected. In addition, it is necessary to ensure that the time point and time period of the collection are representative to obtain comprehensive vehicle flow information. By collecting the vehicle flow values ​​at multiple predetermined time points within a predetermined time period, the traffic flow conditions at different time periods on the road can be obtained, providing basic data for subsequent data analysis and congestion assessment.

[0034] In the vehicle flow time series arrangement module 120, when arranging the vehicle flow values ​​according to the time dimension, it is necessary to ensure the accuracy of the order and interval of the time points. In addition, the storage and processing methods of the data are considered for subsequent time series feature extraction and analysis. Arranging the vehicle flow values ​​according to time as a time series input vector can better reflect the time change trend of the vehicle flow, which is helpful for subsequent time series feature analysis and congestion detection.

[0035] In the vehicle flow time series feature analysis module 130, when performing time series feature extraction, it is necessary to select appropriate feature extraction methods and algorithms to accurately capture the time series features of vehicle flow. At the same time, the dimension and representation of the features are considered for subsequent congestion detection and label determination. By performing time series feature extraction on the vehicle flow time series input vector, important features related to vehicle flow can be extracted from the data, which helps to better understand and analyze the changing laws of vehicle flow and provide a basis for the detection and evaluation of congestion levels.

[0036] In the predetermined road congestion detection module 140, when determining the road congestion level label based on the traffic flow time series characteristics, it is necessary to define the congestion level standard and threshold. At the same time, the impact of other factors on the congestion level is considered, such as road capacity, traffic events, etc. By determining the congestion level label of the predetermined road based on the traffic flow time series characteristics, the traffic conditions of the road can be quickly evaluated, which is helpful for real-time monitoring of traffic congestion and providing timely reference for traffic management and decision-making.

[0037] The collaborative work of these modules can realize the collection of traffic flow data, time series feature analysis and congestion detection on the predetermined roads, thus providing support for urban traffic congestion management and optimization. However, the accuracy and reliability of the modules need to be verified and adjusted in actual applications to ensure their effectiveness.

[0038] In one embodiment of the present application, Figure 2A and Figure 2BAs shown in the figure, smart street lights are intelligent devices that use IoT technology to achieve remote control, monitoring and management of street lights. Smart street lights can integrate meteorological sensors such as temperature sensors, humidity sensors and air pressure sensors. These sensors can monitor the meteorological parameters of the environment in real time, provide weather forecasts and meteorological data, and help urban planning and residents make corresponding decisions. Smart street lights can be equipped with high-definition cameras for real-time monitoring of the safety of roads and public areas. Video surveillance can help monitor traffic flow, detect traffic violations, and provide safety warnings and crime prevention functions. Smart street lights can be equipped with RFID readers to identify and track objects or vehicles. For example, the use of RFID technology in parking lots can realize automatic vehicle identification and billing, providing convenient parking services. Smart street lights can be installed with emergency call buttons for pedestrians to seek help in emergencies. These buttons can be connected to emergency service agencies to provide rapid response and rescue. Smart street lights can realize intelligent lighting functions through devices such as light sensors and motion sensors. According to the ambient light and traffic conditions, smart street lights can automatically adjust the brightness and lighting time to provide energy-saving and comfortable lighting effects. Smart street lights can provide wireless network coverage, providing free public Wi-Fi services for city residents and tourists, making it easier for them to connect to the Internet and use online services. Smart street lights can be equipped with display screens or LED light strips to publish real-time information and city announcements. This information can include traffic conditions, weather forecasts, public events, and emergency notices, providing convenient and safe information services. Smart street lights can integrate electric vehicle charging piles to provide charging services for electric vehicles. Such facilities can promote the popularization of electric vehicles and the development of sustainable transportation.

[0039] The multifunctional application of smart street lights can improve the intelligence level of cities and provide more convenient, safe and sustainable urban services. By integrating various technologies and functions, smart street lights have become an important part of urban infrastructure, contributing to the sustainable development of cities and the quality of life of residents.

[0040] Smart street lights can not only save energy and improve the service life and safety of street lights, but also collect and transmit various data of urban roads, providing valuable information for urban management and services. In order to effectively utilize the data collected by smart street lights, an urban road IoT data management platform based on smart street lights is proposed. The main functions of the platform are as follows:

[0041] Data collection: The platform uses wireless communication technology to receive various data collected by smart street lights in real time, such as street light status, environmental parameters, traffic flow, vehicle type, number of pedestrians, etc.

[0042] Data storage: The platform uses a distributed database system to classify and store data according to time, space and type to ensure data integrity and reliability.

[0043] Data analysis: The platform uses big data analysis technology to count, mine and predict data, and generate various reports and charts to reflect the real-time situation and development trends of urban roads.

[0044] Data display: The platform provides a variety of data display methods, such as web pages, mobile terminals, large screens, etc., to show the results of data analysis to different users and departments, and supports data query, download and sharing.

[0045] Data application: Based on the results of data analysis, the platform provides intelligent decision-making support and optimization suggestions for urban management and services, such as adjusting the brightness of street lights, optimizing traffic signals, improving road facilities, and enhancing public safety.

[0046] The urban road IoT data management platform based on smart street lamps is an innovative urban intelligence solution that can effectively improve the management level and service quality of urban roads, and bring huge social and economic benefits to urban construction and development.

[0047] In response to the above technical problems, the technical concept of this application is to use the Internet of Things technology to deploy sensors and equipment on urban roads to collect and monitor traffic data in real time, providing a large amount of information and decision-making basis for urban traffic management. Among them, smart street lights, as part of the urban road Internet of Things system, can not only provide lighting functions, but also collect traffic data such as vehicle flow in real time through built-in sensors, and introduce data processing and analysis algorithms at the back end to perform time series analysis of vehicle flow values, so as to judge the congestion level of the predetermined road, provide valuable information for urban management and services, and help to realize the monitoring, prediction and optimization of traffic congestion.

[0048] Specifically, in the technical solution of the present application, first, the traffic flow values ​​at multiple predetermined time points within a predetermined time period collected by the smart street lamps deployed on the predetermined road are obtained. Through the traffic flow values ​​collected by the smart street lamps, the traffic flow conditions at different time points within the predetermined time period can be obtained. These data provide an objective description of the road traffic conditions and provide a data basis for the subsequent congestion assessment. Through the traffic flow values ​​at multiple predetermined time points, the changing trend of the traffic flow on the road can be observed, which helps to understand the peak and trough periods of the road, as well as the periodic changes in traffic flow, and thus have a more comprehensive understanding of the congestion situation on the road. The traffic flow time series feature analysis module can use these traffic flow values ​​to extract time series features. By analyzing the time series features of the traffic flow, such as peak value, volatility, periodicity, etc., the changing law of the traffic flow can be better understood, providing a basis for the subsequent congestion detection. Based on the traffic flow time series features, the predetermined road congestion detection module can determine the congestion degree label of the predetermined road. These labels can be used to describe the congestion degree of the road, such as smooth, mild congestion, moderate congestion, and severe congestion levels, to help traffic managers and decision makers evaluate the traffic conditions of the road.

[0049] Obtaining the traffic flow values ​​collected by smart street lights can provide a comprehensive understanding of the traffic conditions on the scheduled roads, providing important reference information for traffic management and decision-making. The determination of congestion level labels can help traffic managers understand the congestion of roads in a timely manner and take corresponding measures to alleviate traffic congestion problems and improve traffic efficiency.

[0050] Congestion labels are determined based on traffic flow data and time series feature analysis results to describe the degree of traffic congestion on the road. These labels can help traffic management departments and drivers understand the real-time traffic conditions on the road and make corresponding decisions.

[0051] Congestion level labels are usually expressed in a discrete manner, and the labels include: No congestion, indicating that there is no traffic congestion on the road and vehicles can travel at normal speeds. Mild congestion, indicating that there is a certain degree of traffic congestion on the road, and the vehicle speed has slightly decreased, but it can still maintain a relatively smooth flow. Moderate congestion, indicating that the degree of traffic congestion on the road is high, the vehicle speed has dropped significantly, and the traffic flow is affected to a certain extent. Severe congestion, indicating that the degree of traffic congestion on the road is very high, the vehicle speed has dropped significantly, and the traffic flow is very slow. Severe congestion means that the traffic congestion on the road has reached an extremely high level, vehicles can hardly travel normally, and traffic flow is seriously obstructed.

[0052] The specific classification of congestion level labels can be adjusted according to actual conditions and needs, and different levels of congestion levels can be determined based on indicators such as vehicle density, speed, and delay time. The use of these labels can help traffic management departments monitor road congestion in real time and take corresponding traffic control measures. At the same time, they can also remind drivers to choose appropriate routes to avoid congested areas, thereby improving traffic efficiency and reducing the impact of traffic congestion on urban operations.

[0053] Next, considering that the traffic flow value has a dynamic change law of time series in the time dimension, in order to be able to analyze and judge the time series change law and trend of the traffic flow value, so as to more accurately detect and judge the congestion level of the predetermined road, in the technical solution of the present application, it is necessary to arrange the traffic flow values ​​of the multiple predetermined time points into a traffic flow time series input vector according to the time dimension, so as to integrate the time series distribution information of the traffic flow value.

[0054] In one embodiment of the present application, the vehicle flow time series feature analysis module 130 includes: a vehicle flow time series upsampling unit, used to pass the vehicle flow time series input vector through an upsampling module based on linear interpolation to obtain an upsampled vehicle flow time series input vector; a vector-image conversion unit, used to input the upsampled vehicle flow time series input vector into a vector-image conversion module to obtain a vehicle flow time series image; and a vehicle flow time series feature extraction unit, used to extract image features from the vehicle flow time series image to obtain the vehicle flow time series feature.

[0055] Among them, the function of the vehicle flow time series upsampling unit is to upsample the given vehicle flow time series input vector, increase the time resolution through a linear interpolation-based method, and obtain a more fine-grained vehicle flow time series input vector, which can more accurately capture the changes in vehicle flow and improve the precision and accuracy of subsequent processing.

[0056] The vector-image conversion unit inputs the upsampled traffic flow time series input vector into the vector-image conversion module to convert it into a traffic flow time series image. By converting the traffic flow time series data into an image form, image processing and computer vision methods can be used to extract richer feature information.

[0057] The traffic flow time series feature extraction unit extracts features from the traffic flow time series image in order to extract meaningful feature representations from the image. These features may include peak values, volatility, periodicity, etc., which are used to describe the time series variation of traffic flow. Through the traffic flow time series feature extraction, we can more deeply understand and analyze the changing trend of traffic flow, and provide useful information for subsequent tasks such as congestion detection and traffic prediction.

[0058] In this way, the traffic flow time series upsampling unit can improve the temporal resolution, increase the detailed information of the traffic flow data, and improve the accuracy and precision of subsequent processing. The vector-image conversion unit converts the traffic flow time series data into image form, and uses image processing and computer vision methods to extract richer features, which helps to more comprehensively describe the time series changes of traffic flow. The traffic flow time series feature extraction unit can extract meaningful feature representations from the traffic flow time series image to help understand and analyze the changing laws of traffic flow and provide important reference information for traffic management and decision-making. These features can be used for tasks such as congestion detection, traffic prediction, and optimization of traffic signal control to improve traffic efficiency and reduce congestion problems.

[0059] Then, in order to improve the ability to capture subtle changes in the time series characteristics of the traffic flow on the predetermined road, in the technical solution of the present application, the traffic flow time series input vector is further passed through an upsampling module based on linear interpolation to obtain an upsampled traffic flow time series input vector, so as to increase the time series density and smoothness of the traffic flow data, thereby facilitating a better subsequent representation of the traffic flow time series characteristics of the predetermined road. It should be understood that through linear interpolation upsampling, the data points in the original traffic flow time series input vector can be interpolated to generate more data points. In this way, it helps to increase the resolution in the time dimension, making the time series changes of the traffic flow more detailed and visible. At the same time, linear interpolation can perform smooth interpolation between sampling points, reduce the impact of noise and mutations, and improve the continuity and stability of the data.

[0060] Furthermore, in order to better extract the time series characteristics of the traffic flow on the predetermined road, in the technical solution of the present application, the upsampled traffic flow time series input vector is further input into a vector-image conversion module to obtain a traffic flow time series image. By converting the upsampled traffic flow time series input vector into an image form, the image data can be processed and analyzed using an image processing and analysis algorithm to more accurately analyze the time series changes and trends of the traffic flow on the predetermined road, thereby determining the degree of congestion on the predetermined road. This is because image data has rich information and can contain spatial and temporal structural features, while vector data cannot directly capture these features. In other words, by converting the upsampled traffic flow time series input vector into a traffic flow time series image, it is beneficial to subsequently capture the traffic flow time series change patterns at different time points and spatial locations. These time series features can better reflect the dynamic changes of traffic flow and provide more accurate information for subsequent data processing and road congestion detection.

[0061] In one embodiment of the present application, the vehicle flow time series feature extraction unit is used to: use a convolutional neural network model with a spatial attention mechanism to obtain a spatially explicit vehicle flow time series feature graph as the vehicle flow time series feature.

[0062] Wherein, the vehicle flow time series feature extraction unit is used to: each layer of the convolutional neural network model using the spatial attention mechanism performs convolution processing, mean pooling processing and nonlinear activation processing on the input data in the forward transfer of the layer so that the last layer of the convolutional neural network model using the spatial attention mechanism outputs the spatially explicit vehicle flow time series feature map, wherein the input of the first layer of the convolutional neural network model using the spatial attention mechanism is the vehicle flow time series image.

[0063] Next, after converting the upsampled traffic flow time series input vector into the traffic flow time series image, a convolutional neural network model with excellent performance in extracting implicit features of the image is used to perform feature mining on the traffic flow time series image. In particular, considering that in the actual process of monitoring the time series changes of traffic flow, more attention should be paid to the time series feature information of traffic flow on the predetermined road in spatial position, and the remaining useless interference features should be filtered out. In view of the fact that the attention mechanism can select the focus position, produce a more discriminative feature representation, and the features after adding the attention module will produce adaptive changes as the network deepens. Therefore, in the technical solution of the present application, the traffic flow time series image is processed by a convolutional neural network model using a spatial attention mechanism, so as to enhance the features of the traffic flow time series features of the predetermined road in the traffic flow time series image based on the spatial position, thereby obtaining the traffic flow time series feature information focused on the spatial position of the predetermined road, that is, the spatially explicit traffic flow time series feature map. It is worth mentioning that here, the image features extracted by the spatial attention reflect the weights of the differences in spatial dimension features, which are used to suppress or enhance the features of different spatial positions, so as to extract the time series feature information of the traffic flow spatially focused on the predetermined road.

[0064] The spatial attention mechanism is a technology used in convolutional neural networks to enhance the model's attention to different spatial locations in the input data. It can help the model automatically learn and focus on the most relevant and meaningful spatial regions in the input data, thereby extracting more discriminative and expressive features.

[0065] In the processing of traffic flow time series images, the spatial attention mechanism can be used to highlight the areas in the image that are related to traffic flow changes and enhance the feature expression of these areas to obtain more significant and distinctive traffic flow time series feature maps.

[0066] Specifically, the convolution layer is used to extract feature representations from the image by applying a series of convolution kernels to capture features of different scales and abstraction levels. The attention mechanism is the core component of spatial attention. Based on the feature map of the input data, weights are assigned to different spatial locations through learning to determine the degree of attention of the model on different locations. Common attention mechanisms include self-attention and channel attention. Activation functions are used to introduce nonlinear transformations and increase the expressiveness of the model. Activation functions include ReLU (Rectified Linear Unit), Sigmoid, and Tanh.

[0067] By using the convolutional neural network model with spatial attention mechanism, we can automatically learn and focus on the areas related to traffic flow changes in the traffic flow time series image, highlight the feature expression of these areas, and obtain the spatially explicit traffic flow time series feature map. Such feature maps can better reflect the spatial distribution and change trend of traffic flow, and provide more valuable information for subsequent tasks such as congestion analysis and traffic prediction.

[0068] In one embodiment of the present application, the predetermined road congestion level detection module 140 includes: a feature distribution optimization unit, which is used to perform Hilbert orthogonal space domain representation decoupling on the spatially manifested traffic flow time series feature vector obtained by expanding the spatially manifested traffic flow time series feature graph to obtain an optimized spatially manifested traffic flow time series feature graph; and a road congestion judgment unit, which is used to pass the optimized spatially manifested traffic flow time series feature graph through a classifier to obtain a classification result, and the classification result is used to represent the congestion level label of the predetermined road.

[0069] In particular, in the technical solution of the present application, after the vehicle flow time series input vector is upsampled based on linear interpolation, it is input into the vector-image conversion module to obtain the vehicle flow time series image, and the distribution of the vehicle flow value in the time dimension can be represented by a multi-dimensional distribution based on the time series subdivision position. In this way, by using the convolutional neural network model of the spatial attention mechanism, the time series multi-dimensional cross-correlation feature representation effect of the spatially manifested vehicle flow time series feature graph for the distribution of the vehicle flow value in the time dimension can be improved. However, considering the uneven distribution of the vehicle flow value in the time series direction, when performing the multi-dimensional distribution representation based on the time series subdivision position, the uneven distribution of the flow value in the time series multi-dimensional distribution space will also be magnified, resulting in the spatially manifested vehicle flow time series feature graph having diversified local feature expressions between each local feature distribution, thereby affecting the generalization effect of the spatially manifested vehicle flow time series feature graph as a whole in the classification domain when the spatially manifested vehicle flow time series feature graph is classified by a classifier, and affecting the accuracy of the classification result.

[0070] Based on this, when the applicant of the present application classifies the spatially explicit vehicle flow time series characteristic diagram, it is preferred that the spatially explicit vehicle flow time series characteristic vector obtained by expanding the spatially explicit vehicle flow time series characteristic diagram, for example, denoted as V, is decoupled by Hilbert orthogonal spatial domain representation, which is expressed as follows: the spatially explicit vehicle flow time series characteristic vector obtained by expanding the spatially explicit vehicle flow time series characteristic diagram is decoupled by Hilbert orthogonal spatial domain representation to obtain the optimized spatially explicit vehicle flow time series characteristic diagram; wherein the optimization formula is:

[0071] V'=V1θV2

[0072]

[0073] V2=IθV1

[0074] Wherein, V is the spatially explicit vehicle flow time series feature vector obtained by expanding the spatially explicit vehicle flow time series feature graph, is the global feature mean of the spatially explicit traffic flow time series feature vector obtained by expanding the spatially explicit traffic flow time series feature graph, ||V||2 is the second norm of the spatially explicit traffic flow time series feature vector obtained by expanding the spatially explicit traffic flow time series feature graph, L is the length of the spatially explicit traffic flow time series feature vector obtained by expanding the spatially explicit traffic flow time series feature graph, and I is a unit vector, θ represents vector subtraction, Cov 1D (·) represents the covariance matrix, and V' is the optimized space-manifested traffic flow time series feature vector obtained by expanding the optimized space-manifested traffic flow time series feature graph.

[0075] Here, the Hilbert orthogonal space domain representation decoupling is used to emphasize the essential domain-specific information in the diversified feature expression of the spatially explicit traffic flow time series feature vector V, that is, by performing domain-invariant characterization of the orthogonal space domain decoupling from the overall domain representation of the spatially explicit traffic flow time series feature vector V based on the Hilbert space metric under the vector self-space metric and the vector self-inner product representation, so as to improve the domain adaptive generalization performance of the spatially explicit traffic flow time series feature vector V in the classification domain, thereby improving the classification effect of the spatially explicit traffic flow time series feature graph. In this way, the congestion level of the road can be judged based on the real-time changes in the traffic flow of the road, thereby providing valuable information for urban management and services, and helping to achieve the monitoring, prediction and optimization of traffic congestion.

[0076] In one embodiment of the present application, the road congestion judgment unit includes: a matrix expansion subunit, used to expand the optimized spatial explicit traffic flow time series feature map into a classification feature vector according to a row vector or a column vector; a fully connected encoding subunit, used to use multiple fully connected layers of the classifier to fully connect the classification feature vector to obtain an encoded classification feature vector; and a classification subunit, used to pass the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

[0077] Then, the optimized spatially visualized traffic flow time series feature graph is passed through a classifier to obtain a classification result, and the classification result is used to represent the congestion level label of the predetermined road. Specifically, in the technical solution of the present application, the classification label of the classifier is the congestion level label of the predetermined road. Therefore, after obtaining the classification result, the congestion level of the predetermined road is judged by using it to provide valuable information for urban management and services, which is helpful to realize the monitoring, prediction and optimization of traffic congestion level.

[0078] The classifier can associate the optimized feature map with the corresponding congestion level label to predict the congestion level of the road. This enables traffic managers to quickly understand the traffic status of the road, including no congestion, light congestion, moderate congestion, heavy congestion and severe congestion levels. By acquiring traffic flow data in real time and generating corresponding feature maps, the traffic conditions of the road can be monitored in real time. Traffic managers can take timely measures based on the classification results, such as adjusting traffic signal timing, guiding traffic flow, etc., to alleviate congestion and improve traffic efficiency. The classification results provide an accurate prediction of the degree of road congestion and can provide decision support for traffic managers. For example, according to different congestion levels, traffic planning can be optimized, road construction and maintenance plans can be adjusted, and traffic signal control strategies can be improved, thereby improving the operating efficiency and travel experience of urban traffic.

[0079] In summary, the urban road IoT data management platform 100 based on the embodiment of the present application is explained, which utilizes IoT technology to deploy sensors and devices on urban roads to collect and monitor traffic data in real time, thereby providing a large amount of information and decision-making basis for urban traffic management. Among them, as part of the urban road IoT system, smart street lights can not only provide lighting functions, but also collect traffic data such as vehicle flow in real time through built-in sensors, and introduce data processing and analysis algorithms at the back end to perform time series analysis of vehicle flow values, so as to judge the congestion level of the predetermined road, provide valuable information for urban management and services, and help to realize the monitoring, prediction and optimization of traffic congestion levels.

[0080] In one embodiment of the present application, Figure 3 This is a flowchart of a method for managing urban road Internet of Things data according to an embodiment of the present application. Figure 4 Schematic diagram of the urban road IoT data management method architecture according to an embodiment of the present application. Figure 3 and Figure 4 As shown, the urban road Internet of Things data management method includes: 210, collecting vehicle flow values ​​at multiple predetermined time points within a predetermined time period by smart street lamps deployed on predetermined roads; 220, arranging the vehicle flow values ​​at the multiple predetermined time points into a vehicle flow time series input vector according to the time dimension; 230, performing time series feature extraction on the vehicle flow time series input vector to obtain a vehicle flow time series feature; 240, determining a congestion level label of a predetermined road based on the vehicle flow time series feature.

[0081] Those skilled in the art will appreciate that the specific operations of each step in the above-mentioned urban road IoT data management method have been referenced above. Figure 1 The description of the urban road Internet of Things data management platform in Figure 2 has been introduced in detail, and therefore, its repeated description will be omitted.

[0082] Figure 5 FIG. 1 is an application scenario diagram of the urban road IoT data management platform according to an embodiment of the present application. Figure 5 As shown, in this application scenario, first, the smart street lamps deployed on the predetermined road collect the traffic flow values ​​at multiple predetermined time points within a predetermined time period (for example, Figure 5 Then, the obtained traffic flow value is input into a server (for example, Figure 5 In S) shown in , the server is capable of processing the traffic flow value based on the urban road Internet of Things data management algorithm to determine the congestion level label of the predetermined road.

[0083] The basic principles of the present application are described above in conjunction with specific embodiments. However, it should be noted that the advantages, strengths, effects, etc. mentioned in the present application are only examples and not limitations, and it cannot be considered that these advantages, strengths, effects, etc. are required by each embodiment of the present application. In addition, the specific details disclosed above are only for the purpose of illustration and ease of understanding, not for limitation, and the above details do not limit the present application to being implemented by adopting the above specific details.

[0084] The above description of the disclosed aspects is provided to enable any person skilled in the art to make or use the present application. Various modifications to these aspects will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other aspects without departing from the scope of the present application. Therefore, the present application is not intended to be limited to the aspects shown herein, but rather to the widest scope consistent with the principles and novel features disclosed herein.

[0085] Finally, it should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or terminal device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or terminal device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or terminal device including the elements.

[0086] The above description has been given for the purpose of illustration and description. In addition, this description is not intended to limit the embodiments of the present application to the forms disclosed herein. Although multiple example aspects and embodiments have been discussed above, those skilled in the art will recognize certain variations, modifications, changes, additions and sub-combinations thereof.

Claims

1. An urban road Internet of Things data management platform, characterized in that: include: A vehicle flow data collection module is used to collect vehicle flow values ​​at multiple predetermined time points within a predetermined time period through smart street lights deployed on predetermined roads; A vehicle flow time series arrangement module, used for arranging the vehicle flow values ​​at the plurality of predetermined time points into a vehicle flow time series input vector according to the time dimension; A vehicle flow time series feature analysis module, used for extracting time series features from the vehicle flow time series input vector to obtain vehicle flow time series features; A predetermined road congestion detection module, used to determine a congestion level label of a predetermined road based on the traffic flow time series characteristics; The vehicle flow time series characteristic analysis module includes: A vehicle flow time series upsampling unit, used for passing the vehicle flow time series input vector through an upsampling module based on linear interpolation to obtain an upsampled vehicle flow time series input vector; A vector-to-image conversion unit, configured to input the upsampled vehicle flow time series input vector into a vector-to-image conversion module to obtain a vehicle flow time series image; A vehicle flow time series feature extraction unit, used for extracting image features from the vehicle flow time series image to obtain the vehicle flow time series feature; The vehicle flow time series feature extraction unit is used to: use the vehicle flow time series image through a convolutional neural network model using a spatial attention mechanism to obtain a spatially explicit vehicle flow time series feature graph as the vehicle flow time series feature; The vehicle flow time series feature extraction unit is used to: perform convolution processing, mean pooling processing and nonlinear activation processing on input data in the forward transfer of each layer of the convolutional neural network model using the spatial attention mechanism, so that the last layer of the convolutional neural network model using the spatial attention mechanism outputs the spatial explicit vehicle flow time series feature map, wherein the input of the first layer of the convolutional neural network model using the spatial attention mechanism is the vehicle flow time series image; The predetermined road congestion degree detection module comprises: A feature distribution optimization unit, used for performing Hilbert orthogonal space domain representation decoupling on the spatially explicit traffic flow time series feature vector obtained by expanding the spatially explicit traffic flow time series feature graph to obtain an optimized spatially explicit traffic flow time series feature graph; The road congestion judgment unit is used to pass the optimized spatially visualized traffic flow time series characteristic diagram through a classifier to obtain a classification result, and the classification result is used to represent a congestion degree label of a predetermined road.

2. The urban road Internet of Things data management platform according to claim 1 is characterized in that: The feature distribution optimization unit is used to: perform Hilbert orthogonal space domain representation decoupling on the spatially explicit vehicle flow time series feature vector obtained by expanding the spatially explicit vehicle flow time series feature graph according to the following optimization formula to obtain the optimized spatially explicit vehicle flow time series feature graph; Wherein, the optimization formula is: Wherein, V is the spatially explicit vehicle flow time series feature vector obtained by expanding the spatially explicit vehicle flow time series feature graph, is the global feature mean of the spatially explicit traffic flow time series feature vector obtained by expanding the spatially explicit traffic flow time series feature graph, ||V||2 is the second norm of the spatially explicit traffic flow time series feature vector obtained by expanding the spatially explicit traffic flow time series feature graph, L is the length of the spatially explicit traffic flow time series feature vector obtained by expanding the spatially explicit traffic flow time series feature graph, and I is a unit vector, Represents vector subtraction, Cov 1D (·) represents the covariance matrix, and V' is the optimized space-manifested traffic flow time series feature vector obtained by expanding the optimized space-manifested traffic flow time series feature graph.

3. The urban road Internet of Things data management platform according to claim 2 is characterized in that: The road congestion determination unit comprises: A matrix expansion subunit, used for expanding the optimized spatial explicit traffic flow time series feature diagram into a classification feature vector according to a row vector or a column vector; a fully connected encoding subunit, configured to perform fully connected encoding on the classification feature vector using a plurality of fully connected layers of the classifier to obtain an encoded classification feature vector; and The classification subunit is used to pass the encoded classification feature vector through the Softmax classification function of the classifier to obtain the classification result.

Images