Live internet edge integrated city vitality heat analysis method and system

Abstract

This invention relates to a method and system for integrated edge-end urban vitality heat analysis using a video network. The method includes: acquiring video stream data and environmental parameters within a target area; detecting target objects in the video stream data using the YOLOv8s-Mobile architecture; and cleaning the environmental parameters. The video stream data and environmental parameters are then encrypted and transmitted to a cloud center. Deep analysis of the video stream data is performed at the cloud center to extract spatiotemporal features, and motion information of target objects is mined based on these features using deep learning algorithms. The environmental parameters, video stream data, and motion information of target objects are fused to obtain fused multi-source data, and a vitality heat model is used to calculate the vitality index within the target area. A heat distribution based on the vitality index is constructed using a multi-dimensional correlation graph, and this heat distribution is sent to the front-end display layer to showcase the heat distribution of the target area.

Description

Technical Field

[0001] This invention relates to the fields of smart city and artificial intelligence technology, and in particular to a method and system for analyzing urban vitality heat maps using a video network-edge integrated approach. Background Technology

[0002] With the deepening of smart city construction globally, urban vitality analysis, as a key component of refined urban management and sustainable development, has received widespread attention. However, traditional urban vitality analysis methods have revealed numerous problems and are no longer sufficient to meet the demands of rapid modern urban development.

[0003] Currently, traditional analytical methods primarily rely on static data, such as Points of Interest (POI) data. This data only reflects the static distribution of urban functions. Population census data has a long update cycle, making it difficult to reflect the dynamic flow of the urban population in real time. In real-world scenarios, urban vitality varies significantly throughout the day and across seasons. For example, commercial areas see high pedestrian traffic during weekday rush hours, while leisure and entertainment areas are more active on weekends. These dynamic changes are difficult to accurately capture using static data, and traditional analytical methods lack support from dynamic perception data such as real-time pedestrian and vehicle traffic and environmental quality. This leads to a certain degree of bias in the assessment of urban vitality, making it difficult to provide more accurate decision-making basis. In video stream target detection and statistics, traditional methods rely heavily on centralized cloud processing. However, the massive amounts of video data generated by numerous cameras in the city face not only network bandwidth bottlenecks during transmission to the cloud but also significant latency. For example, during large-scale events, simultaneous uploading of video data from multiple areas can cause network congestion, resulting in severely delayed video analysis results and making it difficult to meet the needs of real-time urban management. Moreover, the high bandwidth costs resulting from long-term reliance on cloud processing have placed a significant economic burden on urban management departments. Current vitality assessments fail to adequately consider their connection to urban issues such as carbon emissions and traffic congestion. Because a city is a complex ecosystem, its various elements interact with each other. For example, highly vibrant commercial areas are often accompanied by higher carbon emissions and traffic congestion, but existing assessment systems lack fine-grained spatiotemporal mapping models, making it difficult to accurately reveal the intrinsic connections between these elements. This hinders the comprehensive consideration and optimal resource allocation when formulating urban governance strategies.

[0004] In summary, traditional urban vitality heat map analysis methods lack comprehensiveness in assessing urban vitality and suffer from limitations in data support, resulting in poor real-time performance during video data analysis. Furthermore, traditional analytical methods, to some extent, neglect the relationships between the various elements of urban vitality heat map analysis. Summary of the Invention

[0005] Therefore, it is necessary to address the aforementioned technical problems by providing a visual network-based edge-end integrated urban vitality heat analysis method and system that, while lacking comprehensiveness and real-time performance, can integrate the relationships between various elements in urban vitality heat analysis.

[0006] This invention provides a method for analyzing urban vitality heatmaps using a video network-based edge-end integrated approach, the method comprising:

[0007] The system acquires video stream data and environmental parameters within the target area, detects target objects in the video stream data using the YOLOv8s-Mobile architecture, and performs data cleaning on the environmental parameters to remove outliers and noise.

[0008] The video stream data of the identified target object and the cleaned environmental parameters are encrypted to obtain encrypted data, and the encrypted data is transmitted to the cloud center.

[0009] The cloud center performs in-depth analysis on the video stream data to extract spatiotemporal features from the video stream data, and uses deep learning algorithms to mine motion information of the target object based on the spatiotemporal features.

[0010] The environmental parameters, video stream data, and motion information of the target object are fused to obtain fused multi-source data, and the vitality heat model is called to calculate the vitality index within the target area based on the spatiotemporal features and the fused multi-source data;

[0011] A heat distribution based on the vitality index is constructed using a multidimensional correlation graph, and the heat distribution is sent to the front-end display layer to display the heat distribution of the target area;

[0012] The video stream data is collected by cameras installed in the target area, and the environmental parameter is the air quality index obtained by environmental sensors in the target area, which is measured by PM2.5 concentration. The target objects include people and vehicles, and the activity heat model is constructed based on the movement information of the target objects, the commercial transaction volume in the target area, and their respective weights.

[0013] In one embodiment, the steps of acquiring video stream data and environmental parameters within the target area, detecting target objects in the video stream data using the YOLOv8s-Mobile architecture, and cleaning the environmental parameters to remove outliers and noise include:

[0014] The YOLOv8s-Mobile architecture is run on the edge computing box to call the YOLOv8s-Mobile architecture to detect people and vehicles in the video stream data, and to label the identified people and vehicles in order to classify the people and vehicles.

[0015] The DBSCAN algorithm is used to dynamically adjust the clustering radius to eliminate overlapping video streams captured by multiple different cameras within the target area.

[0016] In one embodiment, the process of fusing the environmental parameters, video stream data, and motion information of the target object to obtain fused multi-source data, and then using a vitality heat model to calculate a vitality index within the target area based on the spatiotemporal features and the fused multi-source data, includes:

[0017] The target area is divided into multiple units, and multi-source data such as traffic flow data of people and vehicles, environmental quality index, and commercial transaction volume are extracted from the encrypted data.

[0018] The multi-source data is matched with the corresponding sub-region of each unit, and the multi-source data is fused at the time granularity in every first time period to obtain multi-dimensional fused data corresponding to each unit.

[0019] In this context, the multi-dimensional fused data corresponding to each unit is a subset of the fused multi-source data.

[0020] In one embodiment, the step of fusing the environmental parameters, video stream data, and motion information of the target object to obtain fused multi-source data, and then using a vitality heat model to calculate a vitality index within the target area based on the spatiotemporal features and the fused multi-source data, further includes:

[0021] The flow data of people and vehicles, environmental quality index and commercial transaction volume of the first unit in the target area during the first time period are correlated and analyzed to calculate the vitality index of the first unit during the first time period.

[0022] The first unit is any unit within the target area.

[0023] In one embodiment, the vitality index is calculated as follows:

[0024] ;

[0025] In the formula, H t V represents the vitality index for the first time period. ped V represents the normalized value for pedestrian and vehicle traffic flow. maxThe maximum number of people and vehicles in the first time period is represented by α, where α is the weight of the influence of people and vehicles on the vitality index, and S is the maximum value of the flow of people and vehicles in the first time period. c β represents the commercial transaction volume, and β represents the weight of the commercial transaction volume on the vitality index. PM2.5 t α represents the average PM2.5 concentration during the first time period, γ represents the weight of the PM2.5 concentration on the vitality index, t represents the first time period, and α + β + γ = 1.

[0026] In one embodiment, the step of constructing a heat distribution based on the vitality index using a multidimensional correlation graph and sending the heat distribution to the front-end display layer to display the heat distribution of the target area further includes:

[0027] The spatial clustering of vitality index and air quality index within the target area is analyzed using Moran's I index to determine the spatial correlation between the vitality index and air quality index.

[0028] When the Moran's I index exceeds the first threshold, it is determined that the vitality index and the air quality index are positively correlated spatially.

[0029] When the Moran's I index is lower than the first threshold, it is determined that the vitality index and the air quality index are negatively correlated spatially.

[0030] In one embodiment, the method further includes:

[0031] The congestion index is calculated based on the traffic flow and the speed of people and vehicles within the first unit area. When the vitality index of any unit within the target area exceeds the second threshold and the congestion index exceeds the third threshold, the first instruction is triggered.

[0032] In response to the first instruction, the signal lights in the area where the first unit is located are activated by a filtering coordination mechanism, which is used to adjust the lighting duration and switching sequence of the signal lights.

[0033] This invention also provides a video network edge-end integrated urban vitality heat analysis system, the system comprising:

[0034] The data preprocessing module is used to acquire video stream data and environmental parameters within the target area, detect target objects in the video stream data through the YOLOv8s-Mobile architecture, and perform data cleaning on the environmental parameters to remove outliers and noise.

[0035] The data encryption and transmission module is used to encrypt the video stream data of the target object and the cleaned environmental parameters to obtain encrypted data, and transmit the encrypted data to the cloud center;

[0036] The data analysis module is used to perform in-depth analysis on the video stream data through the cloud center to extract the spatiotemporal features in the video stream data, and to mine the motion information of the target object based on the spatiotemporal features using deep learning algorithms;

[0037] The vitality calculation module is used to fuse the environmental parameters, video stream data and motion information of the target object to obtain fused multi-source data, and call the vitality heat model to calculate the vitality index in the target area based on the spatiotemporal features and the fused multi-source data;

[0038] The heat display module is used to construct a heat distribution based on the vitality index through a multi-dimensional correlation map, and send the heat distribution to the front-end display layer to display the heat distribution of the target area;

[0039] The video stream data is collected by cameras installed in the target area, and the environmental parameter is the air quality index obtained by environmental sensors in the target area, which is measured by PM2.5 concentration. The target objects include people and vehicles, and the activity heat model is constructed based on the movement information of the target objects, the commercial transaction volume in the target area, and their respective weights.

[0040] The present invention also provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the video network edge-end integrated urban vitality heat analysis method as described above.

[0041] The present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, implements the video network edge-end integrated urban vitality heat analysis method as described above.

[0042] The aforementioned video-network edge-device integrated urban vitality heat map analysis method and system acquires video stream data and environmental parameters within a target area. It then uses the YOLOv8s-Mobile architecture to detect target objects in the video stream data and cleans the environmental parameters to remove outliers and noise. The video stream data of the identified target objects and the cleaned environmental parameters are then encrypted to obtain encrypted data, which is transmitted to the cloud center. Subsequently, the cloud center performs deep analysis on the video stream data to extract its spatiotemporal features and uses deep learning algorithms to mine motion information of the target objects based on these features. The environmental parameters, video stream data, and target object motion information are fused to obtain fused multi-source data. A vitality heat map model is then used to calculate the vitality index within the target area based on the spatiotemporal features and the fused multi-source data. Finally, a heat map based on the vitality index is constructed using a multi-dimensional correlation graph, and this heat map is sent to the front-end display layer to showcase the heat map distribution of the target area. This method integrates dynamic data such as video traffic flow, air quality, and business transactions, enabling a more comprehensive analysis of urban vitality by combining the correlations of various factors. The data preprocessing and encrypted transmission process also reduces cloud load to some extent, effectively preventing data congestion and improving the real-time performance of data transmission. Attached Figure Description

[0043] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0044] Figure 1 This is one of the flowcharts of the video network edge-end integrated urban vitality heat analysis method provided by the present invention;

[0045] Figure 2 A schematic diagram of the overall architecture of the urban vitality heat analysis method of the video network edge-end integrated urban vitality heat analysis method provided in a specific embodiment of the present invention;

[0046] Figure 3 This is the second flowchart illustrating the integrated edge-end urban vitality heat analysis method for video networks provided by the present invention.

[0047] Figure 4 The third flowchart illustrates the integrated edge-end urban vitality heat analysis method for video networks provided by this invention.

[0048] Figure 5The fourth flowchart illustrates the integrated edge-end urban vitality heat analysis method for video networks provided by this invention.

[0049] Figure 6 The fifth flowchart illustrates the integrated urban vitality heat analysis method for video network edge devices provided by this invention.

[0050] Figure 7 This is a schematic diagram of the integrated urban vitality heat analysis system for video network edge devices provided by the present invention.

[0051] Figure 8 This is a diagram of the internal structure of the electronic device provided by the present invention. Detailed Implementation

[0052] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0053] The following is combined Figures 1 to 8 This invention describes a method and system for analyzing urban vitality heat maps using a video network that integrates edge and terminal interfaces.

[0054] like Figure 1 As shown, in one embodiment, a method for analyzing urban vitality heatmaps using a video network-based edge-end system includes the following steps:

[0055] Step S110: Obtain video stream data and environmental parameters within the target area, detect target objects in the video stream data using the YOLOv8s-Mobile architecture, and perform data cleaning on the environmental parameters to remove outliers and noise.

[0056] The video stream data is collected by cameras installed in the target area, and the environmental parameters are the air quality index obtained by environmental sensors in the target area. The air quality index is measured by PM2.5 concentration, and the target objects include people and vehicles.

[0057] Specifically, the terminal server acquires video stream data and environmental parameters within the target area, and detects target objects in the video stream data through the YOLOv8s-Mobile architecture. At the same time, it performs data cleaning on the environmental parameters measured by the environmental sensors to remove outliers and noise from the environmental parameters.

[0058] Combination Figure 2As shown in the specific embodiment, the urban vitality heat analysis method integrating the visual network edge and terminal provided by the present invention is implemented through the edge layer, network layer, cloud center, and application layer, and the implementation algorithms of the edge layer, network layer, cloud center, and application layer are as follows:

[0059] graph LR

[0060] A [Edge Layer] --> |Encrypted Transmission| B [Network Layer]

[0061] B --> C [Cloud Center]

[0062] C --> D [Application Layer]

[0063] subgraph edge layer

[0064] A1 (Intersection Camera) --> A2 [Edge Computing Box: Object Detection]

[0065] A3 (Store Camera) --> A2

[0066] A4 (Environmental Sensor) --> A5 [Edge Aggregation Node: Data Cleaning]

[0067] end

[0068] subgraph cloud center

[0069] C1 [Video Stream Analysis Engine] --> C2 [Spatiotemporal Feature Extraction]

[0070] C2 --> C3 [Vitality Heat Model]

[0071] C3 --> C4 [Multidimensional Association Graph]

[0072] end

[0073] subgraph application layer

[0074] D1 [Real-time Heatmap] --> D2 [Governance Decision Platform]

[0075] D3 [Carbon Emission Prediction] --> D2

[0076] end

[0077] In this embodiment, the edge layer serves as the frontline for data acquisition, deploying various types of devices: intersection cameras and storefront cameras are responsible for collecting video stream data. These cameras possess high-definition shooting capabilities, clearly capturing passing people, vehicles, and non-motorized vehicles. Environmental sensors are used to monitor environmental parameters in real time, such as PM2.5 concentration and air quality index. The edge computing box bears the heavy responsibility of target detection, employing an advanced YOLOv8s-Mobile architecture, enabling 16-channel concurrent detection of people / vehicles / non-motorized vehicles on a single device under 1080P video streams, with a detection accuracy of mAP@0.5≥90%. The edge aggregation node cleans the data collected by the environmental sensors, removing outliers and noise to ensure data accuracy and reliability.

[0078] In this embodiment, the YOLOv8s-Mobile architecture is a lightweight object detection model specifically designed for edge devices. Its network structure is optimized to significantly reduce computational load and memory consumption while maintaining detection accuracy. Running this model on an edge computing box enables rapid detection of people, vehicles, and non-motorized vehicles in video stream data.

[0079] For example, at a busy intersection, dozens of target objects may pass by every second. The YOLOv8s - Mobile architecture can accurately identify and classify these target objects in a short time, with a detection accuracy of mAP@0.5≥90%, which can meet the needs of real-time urban monitoring.

[0080] Furthermore, an improved DBSCAN algorithm (dynamic ε-value adjustment) is needed to effectively eliminate duplicate counting across cameras. In practical applications, due to overlapping shooting ranges of different cameras, the same target object may be captured by multiple cameras, leading to duplicate counting. The improved DBSCAN algorithm addresses this issue by dynamically adjusting the clustering radius ε. The DBSCAN algorithm adaptively adjusts the ε value based on the target object's movement speed. When the target object moves quickly, the ε value is appropriately increased to ensure that the trajectories of the same target object across different cameras can be correctly merged; when the target object moves slowly, the ε value is decreased to improve the accuracy of deduplication. In this way, the counting error rate can be controlled to ≤2.3%.

[0081] Step S120: Encrypt the video stream data of the identified target object and the cleaned environmental parameters to obtain encrypted data, and transmit the encrypted data to the cloud center.

[0082] Specifically, the terminal server uses an encryption algorithm to encrypt the video stream data of the identified target object and the cleaned environmental parameters obtained in step S110, obtains encrypted data, and transmits the encrypted data to the cloud center.

[0083] Combination Figure 2 As shown in the specific embodiment, the urban vitality heat analysis method integrating the visual network edge and terminal provided by the present invention involves the network layer being responsible for securely and efficiently transmitting the data collected and preliminarily processed by the edge layer to the cloud center through encrypted transmission. The use of encryption algorithms also ensures data security during transmission, preventing data theft or tampering.

[0084] Step S130: Perform in-depth analysis on the video stream data through the cloud center to extract the spatiotemporal features in the video stream data, and mine the motion information of the target object based on the spatiotemporal features using deep learning algorithms.

[0085] Specifically, the terminal server performs in-depth analysis on the received video stream data through the cloud center to extract the spatiotemporal features in the video stream data, and calls deep learning algorithms to mine the motion information of the target object based on the spatiotemporal features, including the motion trajectory and frequency of appearance of people and vehicles.

[0086] Combination Figure 2 As shown in the specific embodiment, the urban vitality heat analysis method integrating the video network edge and terminal provided by this invention uses a cloud-based video stream analysis engine to perform deep analysis on the video stream data uploaded from the edge layer, extracting spatiotemporal features from the video stream data. The spatiotemporal feature extraction module uses deep learning algorithms to mine information such as the motion trajectory and frequency of occurrence of target objects in the video stream data. Based on the extracted spatiotemporal features and fused multi-source data, the vitality heat model calculates a street-level vitality index. Finally, a multi-dimensional correlation graph construction module establishes a dynamic correlation between vitality, carbon emissions, and traffic congestion, providing more comprehensive information support for urban governance.

[0087] Step S140: The environmental parameters, video stream data and motion information of the target object are fused to obtain fused multi-source data, and the vitality heat model is called to calculate the vitality index in the target area based on the spatiotemporal characteristics and the fused multi-source data.

[0088] The vitality heat model is constructed based on the motion information of the target object, the commercial transaction volume within the target area, and their respective weights.

[0089] Specifically, the terminal server fuses environmental parameters, video stream data, and motion information of the target object to obtain fused multi-source data, and calls the vitality heat model to calculate the vitality index within the target area based on spatiotemporal characteristics and the fused multi-source data.

[0090] Combination Figure 2As shown in the specific embodiment, the urban vitality heat analysis method integrating the edge and terminal of the visual network provided by this invention performs spatiotemporal alignment of pedestrian and vehicle traffic reported by the edge layer, PM2.5 data measured by environmental sensors, and commercial transaction volume recorded by commercial POS machines. Using street blocks as units, data from different sources are spatially matched to ensure that the data corresponds to the accurate street location. In terms of time granularity, a unified 15-minute interval is used, meaning that various types of data are integrated every 15 minutes for subsequent analysis. For example, pedestrian traffic, PM2.5 concentration, and commercial transaction volume of a certain street block within 15 minutes are correlated and analyzed to comprehensively understand the vitality status of that street block during this time period.

[0091] In this embodiment, the expression for calculating the vitality index (the expression for the vitality heat model) is:

[0092] ;

[0093] In the formula, H t V represents the vitality index for the first time period. ped V represents the normalized value for pedestrian and vehicle traffic flow. max The maximum number of people and vehicles in the first time period is represented by α, where α is the weight of the influence of people and vehicles on the vitality index, and S is the maximum value of the flow of people and vehicles in the first time period. c β represents the commercial transaction volume, and β represents the weight of the commercial transaction volume on the vitality index. PM2.5 t The value of PM2.5 concentration is the average value within the first time period, γ represents the weight of PM2.5 concentration on the vitality index, t represents the first time period, and α + β + γ = 1. These weights were determined through extensive experiments and data analysis, comprehensively considering the degree of influence of various factors on urban vitality. For example, in a neighborhood dominated by commercial activities, pedestrian traffic and commercial transaction volume have a greater impact on vitality, while environmental quality has a relatively smaller impact. By setting appropriate weights, the vitality index of this neighborhood can be calculated more accurately.

[0094] In this embodiment, Moran's I index is also used to analyze the spatial clustering of vitality and carbon emissions. Moran's I index is a statistical indicator used to measure spatial autocorrelation, with a value range of [-1, 1]. When Moran's I index is greater than 0, it indicates a positive spatial correlation between the two, meaning that high-vitality areas tend to be adjacent to other high-vitality areas, and high-carbon-emission areas tend to be adjacent to other high-carbon-emission areas. When Moran's I index is less than 0, it indicates a negative spatial correlation. When Moran's I index is close to 0, it indicates that the spatial distribution of the two is random. By calculating the Moran's I index, areas with high vitality and high emissions conflict can be identified, such as certain commercial areas that are high in vitality but also high in carbon emissions, providing key areas of concern for urban governance.

[0095] In this embodiment, when the street activity value > threshold and the congestion index > 0.7, the green wave coordination of traffic lights at surrounding intersections is triggered. The activity value is calculated using the activity index, while the congestion index is calculated using data such as traffic flow and vehicle speed. When the above conditions are met, it indicates that the street has high activity and severe traffic congestion. At this time, the green wave coordination mechanism of traffic lights needs to be activated. By adjusting the duration and switching sequence of traffic lights, vehicles can pass through multiple intersections continuously, thereby reducing waiting time, improving road traffic efficiency, and alleviating traffic congestion.

[0096] Step S150: Construct a heat distribution based on the vitality index through a multidimensional correlation map, and send the heat distribution to the front-end display layer to display the heat distribution of the target area.

[0097] Specifically, the terminal server constructs a heat distribution based on the vitality index through a multi-dimensional correlation graph, and sends the constructed heat distribution to the front-end display layer to visualize the heat distribution of the target area.

[0098] Combination Figure 2 As shown in the specific embodiment, the urban vitality heat analysis method integrating video network edge and terminal provided by this invention uses the application layer to intuitively display the distribution of vitality heat in different areas of the city in a real-time heat map. Users can clearly see which areas have high vitality heat and which areas have low vitality heat. Based on the vitality heat analysis results and multi-dimensional correlation maps, the governance decision-making platform provides decision-making basis for city managers, such as formulating traffic management plans and optimizing commercial layouts. The carbon emission prediction module uses the established correlation model to predict the city's future carbon emissions, helping the city achieve its energy conservation and emission reduction goals.

[0099] In this embodiment, a commercial area is equipped with 12 intersection cameras (each equipped with a target detection module) and 8 air quality sensors. The intersection cameras capture pedestrians, vehicles, and non-motorized vehicles in real time, and the target detection modules perform target detection based on the YOLOv8s-Mobile architecture. The air quality sensors continuously monitor the air quality within the commercial area, such as PM2.5 concentration and sulfur dioxide concentration. These devices transmit the collected data to an edge computing box for preliminary processing in real time.

[0100] Cloud-trained model: A vitality-average transaction value regression model fitted based on historical data (R² = 0.76). By collecting vitality index data and corresponding average transaction value data for this business district over a past period, a model was constructed using regression analysis. This model reflects the linear relationship between the vitality index and average transaction value. R² = 0.76 indicates a good fit to the data, and the model can be used to predict changes in average transaction value under different vitality states, providing a reference for business decisions.

[0101] Dynamic Adjustment: When the weekend vitality index exceeds the 85th percentile, the free parking period for nearby parking lots is automatically extended to alleviate congestion. The vitality value is calculated using the vitality index, and the 85th percentile indicates that 85% of the vitality index data is below this value. When the weekend vitality value exceeds this threshold, it indicates that the business district is very active, which may lead to traffic congestion. At this time, the system automatically extends the free parking period for nearby parking lots to attract vehicles to park, thereby reducing the number of vehicles on the road, effectively alleviating congestion, and improving the overall operational efficiency of the business district.

[0102] Identifying high-activity, high-carbon-emission neighborhoods (Moran's I > 0.4): Spatial autocorrelation analysis of the activity and carbon emissions of various urban neighborhoods is performed by calculating the Moran's I index. When Moran's I > 0.4, it indicates that the neighborhood exhibits spatial characteristics of high activity and high carbon emissions. For example, some mixed industrial and commercial areas, due to the dual impact of industrial production and commercial activities, have high activity but also high carbon emissions.

[0103] The coordinated traffic light system adds a bus priority signal phase: For identified high-activity, high-carbon-emission neighborhoods, the coordinated traffic light system adds a bus priority signal phase. In the intersection traffic light settings of these neighborhoods, the green light time for buses is appropriately extended to ensure buses can quickly pass through intersections, improving the efficiency of public transportation. This not only helps alleviate traffic congestion but also encourages more people to choose public transportation, reducing the use of private cars and thus lowering carbon emissions.

[0104] The aforementioned video-network edge-device integrated urban vitality heat map analysis method acquires video stream data and environmental parameters within the target area. It then uses the YOLOv8s-Mobile architecture to detect target objects in the video stream data and cleans the environmental parameters to remove outliers and noise. The video stream data of the identified target objects and the cleaned environmental parameters are then encrypted, resulting in encrypted data, which is transmitted to the cloud center. Next, the cloud center performs deep analysis on the video stream data to extract its spatiotemporal features and uses deep learning algorithms to mine motion information of the target objects based on these features. The environmental parameters, video stream data, and target object motion information are fused to obtain fused multi-source data. A vitality heat map model is then used to calculate the vitality index within the target area based on the spatiotemporal features and the fused multi-source data. Finally, a heat map based on the vitality index is constructed using a multi-dimensional correlation graph, and this heat map is sent to the front-end display layer to showcase the heat map distribution of the target area. This method integrates dynamic data such as video traffic flow, air quality, and business transactions, enabling a more comprehensive analysis of urban vitality by combining the correlations of various factors. The data preprocessing and encrypted transmission process also reduces cloud load to some extent, effectively preventing data congestion and improving the real-time performance of data transmission.

[0105] like Figure 3 As shown, in one embodiment, the urban vitality heat analysis method integrating video network edge and terminal provided by the present invention includes the following steps in step S110:

[0106] Step S111: Run the YOLOv8s-Mobile architecture on the edge computing box to call the YOLOv8s-Mobile architecture to detect people and vehicles in the video stream data, and label the identified people and vehicles to classify them.

[0107] Step S112: The clustering radius is dynamically adjusted using the DBSCAN algorithm to eliminate overlapping video streams collected by multiple different cameras within the target area.

[0108] like Figure 4 As shown, in one embodiment, the urban vitality heat analysis method integrating video network edge and terminal provided by the present invention includes the following steps in step S140:

[0109] Step S141: Divide the target area into multiple units and extract multi-source data from the encrypted data, including traffic flow data of people and vehicles, environmental quality index, and commercial transaction volume.

[0110] Step S142: Match the multi-source data with the corresponding sub-region of each unit, and fuse the multi-source data at the time granularity of each first time period to obtain the multi-dimensional fused data corresponding to each unit.

[0111] In this context, the multi-dimensional fused data corresponding to each unit is a subset of the fused multi-source data.

[0112] Step S143: Perform correlation analysis on the traffic flow data of people and vehicles, environmental quality index and commercial transaction volume of the first unit in the target area during the first time period to calculate the vitality index of the first unit during the first time period.

[0113] The first unit is any unit within the target area.

[0114] In some embodiments, the urban vitality heat analysis method integrating video network edge and terminal provided by the present invention uses the following expression to calculate the vitality index:

[0115] ;

[0116] In the formula, H t V represents the vitality index for the first time period. ped V represents the normalized value for pedestrian and vehicle traffic flow. max The maximum number of people and vehicles in the first time period is represented by α, where α is the weight of the influence of people and vehicles on the vitality index, and S is the maximum value of the flow of people and vehicles in the first time period. c β represents the commercial transaction volume, and β represents the weight of the commercial transaction volume on the vitality index. PM2.5 t α represents the average PM2.5 concentration during the first time period, γ represents the weight of the PM2.5 concentration on the vitality index, t represents the first time period, and α + β + γ = 1.

[0117] like Figure 5 As shown, in one embodiment, the urban vitality heat analysis method integrating video network edge and terminal provided by the present invention further includes the following steps in step S140:

[0118] Step S510: Analyze the spatial clustering of vitality index and air quality index within the target area using Moran's I index to determine the spatial correlation between vitality index and air quality index.

[0119] Step S520: When Moran's I index exceeds the first threshold, it is determined that the vitality index and the air quality index are spatially positively correlated.

[0120] Step S530: When Moran's I index is below the first threshold, it is determined that the vitality index and the air quality index are negatively correlated spatially.

[0121] like Figure 6 As shown, in one embodiment, the video network edge-end integrated urban vitality heat analysis method provided by the present invention further includes the following steps:

[0122] Step S610: Calculate the congestion index based on the traffic flow and speed of people and vehicles in the first unit area, and trigger the first instruction when the vitality index of any unit in the target area exceeds the second threshold and the congestion index exceeds the third threshold.

[0123] Step S620: In response to the first instruction, the signal lights in the area where the first unit is located are activated by a filtering coordination mechanism. The filtering coordination mechanism is used to adjust the lighting duration and switching sequence of the signal lights.

[0124] The following describes the integrated urban vitality heat analysis system of the video network edge-end provided by the present invention. The integrated urban vitality heat analysis system of the video network edge-end described below and the integrated urban vitality heat analysis method of the video network edge-end described above can be referred to in correspondence.

[0125] like Figure 7 As shown in one embodiment, a video network edge-end integrated urban vitality heat analysis system includes a data preprocessing module 710, a data encryption and transmission module 720, a data analysis module 730, a vitality calculation module 740, and a heat display module 750.

[0126] The data preprocessing module 710 is used to acquire video stream data and environmental parameters within the target area, detect target objects in the video stream data through the YOLOv8s-Mobile architecture, and perform data cleaning on the environmental parameters to remove outliers and noise.

[0127] The data encryption and transmission module 720 is used to encrypt the video stream data of the identified target object and the cleaned environmental parameters to obtain encrypted data, and transmit the encrypted data to the cloud center.

[0128] The data analysis module 730 is used to perform in-depth analysis of video stream data through the cloud center to extract spatiotemporal features from the video stream data, and to mine motion information of target objects based on spatiotemporal features using deep learning algorithms.

[0129] The vitality calculation module 740 is used to fuse environmental parameters, video stream data and motion information of target objects to obtain fused multi-source data, and call the vitality heat model to calculate the vitality index in the target area based on spatiotemporal features and fused multi-source data.

[0130] The heat display module 750 is used to construct a heat distribution based on the vitality index through a multi-dimensional correlation map, and send the heat distribution to the front-end display layer to display the heat distribution of the target area.

[0131] The video stream data is collected by cameras installed in the target area, and the environmental parameters are the air quality index obtained by environmental sensors in the target area. The air quality index is measured by PM2.5 concentration. The target objects include people and vehicles. The activity heat model is constructed based on the movement information of the target objects, the commercial transaction volume in the target area and their respective weights.

[0132] In this embodiment, the data preprocessing module 710 of the video network edge-end integrated urban vitality heat analysis system provided by the present invention is specifically used for:

[0133] Run the YOLOv8s-Mobile architecture on the edge computing box to detect people and vehicles in the video stream data, and label the identified people and vehicles for classification.

[0134] The DBSCAN algorithm is used to dynamically adjust the clustering radius to eliminate overlapping video streams captured by multiple different cameras within the target area.

[0135] In this embodiment, the vitality calculation module 740 of the video network edge-end integrated urban vitality heat analysis system provided by the present invention is specifically used for:

[0136] The target area was divided into multiple units, and multi-source data, including traffic flow data for people and vehicles, environmental quality index, and commercial transaction volume, were extracted from the encrypted data.

[0137] Multi-source data is matched with the corresponding sub-region of each unit, and multi-source data is fused at the time granularity of each first time period to obtain multi-dimensional fused data corresponding to each unit.

[0138] In this context, the multi-dimensional fused data corresponding to each unit is a subset of the fused multi-source data.

[0139] In this embodiment, the vitality calculation module 740 of the video network edge-end integrated urban vitality heat analysis system provided by the present invention is further used for:

[0140] The activity index of the first unit within the target area is calculated by performing correlation analysis on the traffic flow data of people and vehicles, environmental quality index, and commercial transaction volume in the first time period.

[0141] The first unit is any unit within the target area.

[0142] In this embodiment, the urban vitality heat analysis system integrating video network edge and terminal provided by the present invention calculates the vitality index using the following expression:

[0143] ;

[0144] In the formula, H t V represents the vitality index for the first time period. ped V represents the normalized value for pedestrian and vehicle traffic flow. max The maximum number of people and vehicles in the first time period is represented by α, where α is the weight of the influence of people and vehicles on the vitality index, and S is the maximum value of the flow of people and vehicles in the first time period. c β represents the commercial transaction volume, and β represents the weight of the commercial transaction volume on the vitality index. PM2.5 t α represents the average PM2.5 concentration during the first time period, γ represents the weight of the PM2.5 concentration on the vitality index, t represents the first time period, and α + β + γ = 1.

[0145] In this embodiment, the video network edge-end integrated urban vitality heat analysis system provided by the present invention further includes a spatial clustering analysis module, used for:

[0146] The spatial clustering of vitality index and air quality index within the target area was analyzed using Moran's I index to determine the spatial correlation between the two indices.

[0147] When Moran's I index exceeds the first threshold, it is determined that the vitality index and the air quality index are spatially positively correlated.

[0148] When Moran's I index is below the first threshold, it is determined that the vitality index and the air quality index are spatially negatively correlated.

[0149] In this embodiment, the video network edge-end integrated urban vitality heat analysis system provided by the present invention further includes a traffic light filtering coordination module, used for:

[0150] The congestion index is calculated based on the traffic flow and the speed of people and vehicles within the first unit area. When the vitality index of any unit within the target area exceeds the second threshold and the congestion index exceeds the third threshold, the first instruction is triggered.

[0151] In response to the first command, the signal lights in the area where the first unit is located are activated by a filtering coordination mechanism. The filtering coordination mechanism is used to adjust the lighting duration and switching sequence of the signal lights.

[0152] Figure 8 This example illustrates a schematic diagram of the physical structure of an electronic device, which can be a smart terminal. Its internal structure diagram can be as follows: Figure 8As shown. The electronic device includes a processor, internal memory, and a network interface connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The network interface is used to communicate with external terminals via a network connection. When the computer program is executed by the processor, it implements a video network edge-end integrated urban vitality heat analysis method, which includes:

[0153] The system acquires video stream data and environmental parameters within the target area, detects target objects in the video stream data using the YOLOv8s-Mobile architecture, and performs data cleaning on the environmental parameters to remove outliers and noise.

[0154] The video stream data of the identified target object and the cleaned environmental parameters are encrypted to obtain encrypted data, and the encrypted data is transmitted to the cloud center.

[0155] The video stream data is analyzed in depth through the cloud center to extract the spatiotemporal features in the video stream data, and the motion information of the target object is mined based on the spatiotemporal features through deep learning algorithms.

[0156] The environmental parameters, video stream data, and motion information of the target object are fused to obtain fused multi-source data. The vitality heat model is then used to calculate the vitality index within the target area based on spatiotemporal features and the fused multi-source data.

[0157] A heat distribution based on the vitality index is constructed by using a multidimensional correlation map, and the heat distribution is sent to the front-end display layer to display the heat distribution of the target area;

[0158] The video stream data is collected by cameras installed in the target area, and the environmental parameters are the air quality index obtained by environmental sensors in the target area. The air quality index is measured by PM2.5 concentration. The target objects include people and vehicles. The activity heat model is constructed based on the movement information of the target objects, the commercial transaction volume in the target area and their respective weights.

[0159] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present invention and does not constitute a limitation on the electronic device to which the present invention is applied. A specific electronic device may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0160] On the other hand, the present invention also provides a computer storage medium storing a computer program, which, when executed by a processor, implements a method for analyzing urban vitality heatmaps using a video network-based edge-end integrated system. This method includes:

[0161] The system acquires video stream data and environmental parameters within the target area, detects target objects in the video stream data using the YOLOv8s-Mobile architecture, and performs data cleaning on the environmental parameters to remove outliers and noise.

[0162] The video stream data of the identified target object and the cleaned environmental parameters are encrypted to obtain encrypted data, and the encrypted data is transmitted to the cloud center.

[0163] The video stream data is analyzed in depth through the cloud center to extract the spatiotemporal features in the video stream data, and the motion information of the target object is mined based on the spatiotemporal features through deep learning algorithms.

[0164] The environmental parameters, video stream data, and motion information of the target object are fused to obtain fused multi-source data. The vitality heat model is then used to calculate the vitality index within the target area based on spatiotemporal features and the fused multi-source data.

[0165] A heat distribution based on the vitality index is constructed by using a multidimensional correlation map, and the heat distribution is sent to the front-end display layer to display the heat distribution of the target area;

[0166] The video stream data is collected by cameras installed in the target area, and the environmental parameters are the air quality index obtained by environmental sensors in the target area. The air quality index is measured by PM2.5 concentration. The target objects include people and vehicles. The activity heat model is constructed based on the movement information of the target objects, the commercial transaction volume in the target area and their respective weights.

[0167] On another front, a computer program product or computer program is provided, comprising computer instructions stored in a computer-readable storage medium. A processor of an electronic device reads the computer instructions from the computer-readable storage medium, and when the processor executes the computer instructions, it implements a video network edge-device integrated urban vitality heat analysis method, which includes:

[0168] The system acquires video stream data and environmental parameters within the target area, detects target objects in the video stream data using the YOLOv8s-Mobile architecture, and performs data cleaning on the environmental parameters to remove outliers and noise.

[0169] The video stream data of the identified target object and the cleaned environmental parameters are encrypted to obtain encrypted data, and the encrypted data is transmitted to the cloud center.

[0170] The video stream data is analyzed in depth through the cloud center to extract the spatiotemporal features in the video stream data, and the motion information of the target object is mined based on the spatiotemporal features through deep learning algorithms.

[0171] The environmental parameters, video stream data, and motion information of the target object are fused to obtain fused multi-source data. The vitality heat model is then used to calculate the vitality index within the target area based on spatiotemporal features and the fused multi-source data.

[0172] A heat distribution based on the vitality index is constructed by using a multidimensional correlation map, and the heat distribution is sent to the front-end display layer to display the heat distribution of the target area;

[0173] The video stream data is collected by cameras installed in the target area, and the environmental parameters are the air quality index obtained by environmental sensors in the target area. The air quality index is measured by PM2.5 concentration. The target objects include people and vehicles. The activity heat model is constructed based on the movement information of the target objects, the commercial transaction volume in the target area and their respective weights.

[0174] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided by this invention can include non-volatile and / or volatile memory. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or external cache memory.

[0175] By way of illustration and not limitation, RAM is available in a variety of forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), RAMbus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0176] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0177] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. A method for analyzing urban vitality heatmaps using a video network-based edge-end integrated approach, characterized in that: The method includes: The system acquires video stream data and environmental parameters within the target area, detects target objects in the video stream data using the YOLOv8s-Mobile architecture, and performs data cleaning on the environmental parameters to remove outliers and noise. The video stream data of the identified target object and the cleaned environmental parameters are encrypted to obtain encrypted data, and the encrypted data is transmitted to the cloud center. The cloud center performs in-depth analysis on the video stream data to extract spatiotemporal features from the video stream data, and uses deep learning algorithms to mine motion information of the target object based on the spatiotemporal features. The environmental parameters, video stream data, and motion information of the target object are fused to obtain fused multi-source data, and the vitality heat model is called to calculate the vitality index within the target area based on the spatiotemporal features and the fused multi-source data; A heat distribution based on the vitality index is constructed by using a multidimensional correlation graph, and the heat distribution is sent to the front-end display layer to display the heat distribution of the target area; The video stream data is collected by cameras installed in the target area, and the environmental parameter is the air quality index obtained by environmental sensors in the target area, which is measured by PM2.5 concentration. The target objects include people and vehicles, and the activity heat model is constructed based on the movement information of the target objects, the commercial transaction volume in the target area, and their respective weights.

2. The method for analyzing urban vitality heatmaps using a video network-edge integrated system according to claim 1, characterized in that, The process of acquiring video stream data and environmental parameters within the target area, detecting target objects in the video stream data using the YOLOv8s-Mobile architecture, and cleaning the environmental parameters to remove outliers and noise includes: The YOLOv8s-Mobile architecture is run on the edge computing box to call the YOLOv8s-Mobile architecture to detect people and vehicles in the video stream data, and to label the identified people and vehicles in order to classify the people and vehicles. The DBSCAN algorithm is used to dynamically adjust the clustering radius to eliminate overlapping video streams captured by multiple different cameras within the target area.

3. The method for analyzing urban vitality heatmaps using a video network-edge integrated system according to claim 1, characterized in that, The process of fusing the environmental parameters, video stream data, and motion information of the target object to obtain fused multi-source data, and then using a vitality heat model to calculate the vitality index within the target area based on the spatiotemporal features and the fused multi-source data, includes: The target area is divided into multiple units, and multi-source data such as traffic flow data of people and vehicles, environmental quality index, and commercial transaction volume are extracted from the encrypted data. The multi-source data is matched with the corresponding sub-region of each unit, and the multi-source data is fused at the time granularity in every first time period to obtain multi-dimensional fused data corresponding to each unit. In this context, the multi-dimensional fused data corresponding to each unit is a subset of the fused multi-source data.

4. The method for analyzing urban vitality heatmaps using a video network-edge integrated system according to claim 3, characterized in that, The process of fusing the environmental parameters, video stream data, and motion information of the target object to obtain fused multi-source data, and then using a vitality heat model to calculate the vitality index within the target area based on the spatiotemporal features and the fused multi-source data, further includes: The flow data of people and vehicles, environmental quality index and commercial transaction volume of the first unit in the target area during the first time period are correlated and analyzed to calculate the vitality index of the first unit during the first time period. The first unit is any unit within the target area.

5. The method for analyzing urban vitality heatmaps using a video network-edge integrated system according to claim 4, characterized in that, The formula for calculating the vitality index is as follows: ； In the formula, H t V represents the vitality index for the first time period. ped V represents the normalized value for pedestrian and vehicle traffic flow. max The maximum number of people and vehicles in the first time period is represented by α, where α is the weight of the influence of people and vehicles on the vitality index, and S is the maximum value of the flow of people and vehicles in the first time period. c β represents the commercial transaction volume, and β represents the weight of the commercial transaction volume on the vitality index. PM2.5 t α represents the average PM2.5 concentration during the first time period, γ represents the weight of the PM2.5 concentration on the vitality index, t represents the first time period, and α + β + γ = 1.

6. The method for analyzing urban vitality heatmaps using a video network-edge integrated system according to claim 5, characterized in that, The step of constructing a heat distribution based on the vitality index using a multidimensional correlation graph and sending the heat distribution to the front-end display layer to display the heat distribution of the target area also includes: The spatial clustering of vitality index and air quality index within the target area is analyzed using Moran's I index to determine the spatial correlation between the vitality index and air quality index. When the Moran's I index exceeds the first threshold, it is determined that the vitality index and the air quality index are positively correlated spatially. When the Moran's I index is lower than the first threshold, it is determined that the vitality index and the air quality index are negatively correlated spatially.

7. The method for analyzing urban vitality heatmaps using a video network-edge integrated system according to claim 6, characterized in that, The method further includes: The congestion index is calculated based on the traffic flow and the speed of people and vehicles within the first unit area. When the vitality index of any unit within the target area exceeds the second threshold and the congestion index exceeds the third threshold, the first instruction is triggered. In response to the first instruction, the signal lights in the area where the first unit is located are activated by a filtering coordination mechanism, which is used to adjust the lighting duration and switching sequence of the signal lights.

8. A video network edge-end integrated urban vitality heat analysis system, characterized in that, The system includes: The data preprocessing module is used to acquire video stream data and environmental parameters within the target area, detect target objects in the video stream data through the YOLOv8s-Mobile architecture, and perform data cleaning on the environmental parameters to remove outliers and noise. The data encryption and transmission module is used to encrypt the video stream data of the target object and the cleaned environmental parameters to obtain encrypted data, and transmit the encrypted data to the cloud center; The data analysis module is used to perform in-depth analysis on the video stream data through the cloud center to extract the spatiotemporal features in the video stream data, and to mine the motion information of the target object based on the spatiotemporal features using deep learning algorithms; The vitality calculation module is used to fuse the environmental parameters, video stream data and motion information of the target object to obtain fused multi-source data, and call the vitality heat model to calculate the vitality index in the target area based on the spatiotemporal features and the fused multi-source data; The heat display module is used to construct a heat distribution based on the vitality index through a multi-dimensional correlation graph, and send the heat distribution to the front-end display layer to display the heat distribution of the target area; The video stream data is collected by cameras installed in the target area, and the environmental parameter is the air quality index obtained by environmental sensors in the target area, which is measured by PM2.5 concentration. The target objects include people and vehicles, and the activity heat model is constructed based on the movement information of the target objects, the commercial transaction volume in the target area, and their respective weights.

9. An electronic device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 7.

10. A computer storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 7.

Images