A walking street comprehensive management system, method, storage medium and computer device

Abstract

The application provides a walking street comprehensive management system, method, storage medium and computer equipment. Through the collaborative architecture of the perception layer device, the edge computing node, the cloud big data platform and the application service layer, the closed-loop intelligent control of the walking street management is realized. The perception layer device constructs a real-time data acquisition network with global coverage, ensuring accurate capture of multidimensional information such as people flow, environment and vehicles. The edge computing node realizes millisecond-level abnormal event response through the localized people flow analysis and early warning engine, avoiding the treatment lag caused by cloud transmission delay. The cloud big data platform deeply mines data value through the business intelligence module, provides scientific basis for shop location selection and format adjustment, and optimizes the prediction model by training historical data, thereby improving the forward-looking nature of management decisions. The application service layer provides personalized navigation and recommendation services for tourists, and also provides a visual management tool for management personnel.

Description

Technical Field

[0001] This application relates to the field of smart city management technology, and in particular to a pedestrian street integrated management system, method, storage medium and computer equipment. Background Technology

[0002] As a core space integrating urban commerce and leisure, pedestrian streets involve multi-dimensional needs in their operation and management, including crowd control, environmental monitoring, merchant services, and security. However, traditional management models rely on manual inspections and decentralized equipment data collection, which have significant limitations: On the one hand, manual inspections have limited coverage and slow response times, making it difficult to perceive dynamic changes in pedestrian density and environmental conditions across the entire pedestrian street in real time. For example, during peak holiday periods, crowds can easily lead to congestion and even safety hazards, yet early warnings and rapid crowd control are impossible. On the other hand, various management subsystems (such as security monitoring, parking management, and merchant marketing) operate independently, resulting in significant data silos. This hinders collaborative analysis of multi-source data to support accurate decision-making and makes it difficult to provide integrated intelligent services to visitors, leading to low management efficiency and a poor user experience. Furthermore, existing systems lack depth in mining commercial data and cannot combine information such as pedestrian flow characteristics and consumption behavior to provide a scientific basis for merchant operations and business optimization, thus restricting the enhancement of the commercial value of pedestrian streets.

[0003] Therefore, there is an urgent need to build a comprehensive management system that integrates perception, analysis, decision-making, and service to break through the bottlenecks of the traditional model and achieve intelligent, refined, and efficient operation of pedestrian streets. Summary of the Invention

[0004] The purpose of this application is to at least address one of the aforementioned technical deficiencies, particularly the technical deficiency that the traditional management model of pedestrian streets in the prior art cannot achieve intelligent, refined, and efficient operation of pedestrian streets.

[0005] This application provides a pedestrian street integrated management system, the system comprising: The sensing layer devices are deployed in the physical space of the pedestrian street to collect multi-source environmental data in real time; Edge computing nodes are communicatively connected to the perception layer devices and are used to process and analyze the multi-source environmental data in real time to generate preliminary analysis results. The cloud-based big data platform communicates with the edge computing nodes to receive and store the primary analysis results and historical data, and to perform deep data analysis, model training, and business intelligence decision-making. The application service layer communicates with the cloud-based big data platform and provides intelligent application services to management terminals and visitor terminals based on in-depth data analysis results, thereby forming a closed-loop control of the pedestrian street operation.

[0006] Optionally, the sensing layer device includes: The AI ​​video acquisition unit includes multiple high-definition AI cameras deployed at the main entrances and exits of the pedestrian street, key nodes and commercial clusters. Each AI camera is distributed according to a preset density and is used to acquire continuous video streams. The customer flow sensing unit includes customer flow statistics sensors deployed at shop entrances and public areas to collect signals of the number of people passing through and entering designated areas; The vehicle sensing unit includes parking space detectors deployed in each parking lot to collect parking space occupancy status signals; An environmental sensing unit, including a light intensity sensor, is used to collect ambient light data.

[0007] Optionally, the perception layer device includes at least an AI video acquisition unit, and the edge computing node includes a people flow analysis engine for performing the following operations: Receive a continuous video stream of the target area sent by the AI ​​video acquisition unit; Multiple pedestrians in the continuous video stream are identified using a pre-set target detection model, and a multi-target tracking algorithm is called to assign a unique ID to each pedestrian and record the movement trajectory and speed of each pedestrian to obtain the tracking results; Based on the tracking results, the distribution of the number of people, inbound and outbound traffic, crowd density, average movement speed, and dwell time in the target area is calculated and output in real time.

[0008] Optionally, the edge computing node further includes an early warning response engine for: Receive the crowd density output by the crowd flow analysis engine; The population density is compared with a preset density grading threshold, and the comparison result is obtained. The density grading threshold includes at least a first density threshold and a second density threshold. When the comparison result indicates that the population density does not exceed the first density threshold, the target area continues to be monitored by the AI ​​video acquisition unit. When the comparison result indicates that the population density exceeds the first density threshold but does not exceed the second density threshold, a first warning instruction is generated and sent to the application service layer. When the comparison result indicates that the population density exceeds the second density threshold, a second early warning instruction and an emergency response instruction are generated and sent.

[0009] Optionally, the perception layer device includes at least a passenger flow perception unit, and the cloud-based big data platform includes a business intelligence analysis module, which includes: The store value assessment submodule is used to calculate the value index of each store based on historical and real-time foot traffic data and store location attributes through a weighted multi-factor model, and generate a dynamically updated store value map. The merchant operation analysis submodule is used to analyze the merchant's store entry rate and customer dwelling behavior based on the data collected by the customer flow perception unit and the authorized video data in the store, and to generate periodic operation diagnosis reports. The business mix optimization submodule is used to analyze the overall business distribution and customer consumption path of the pedestrian street, identify the complementarity and conflict of business types, and provide suggestions for business type adjustment.

[0010] Optionally, the calculation formula for the weighted multi-factor model is: Store Value Index = W1 × Average Daily Customer Traffic + W2 × Customer Dwell Time + W3 × Complementarity Score of Surrounding Businesses + W4 × Accessibility Score + W5 × Visibility Score; Among them, W1, W2, W3, W4, and W5 are the weight coefficients corresponding to the average daily passenger flow, passenger dwell time, complementaryness score of surrounding business formats, accessibility score, and visibility score, respectively, and W1+W2+W3+W4+W5=1.

[0011] Optionally, the application service layer includes: The intelligent guidance service module is used to provide tourists with optimal route planning based on real-time location and destination, AR real-scene navigation, barrier-free route planning, personalized shop recommendations, and intelligent parking guidance services. The management cockpit module is used to provide the management terminal with a visually integrated dashboard that includes real-time passenger flow heat maps, early warning information, merchant rankings, facility status, and energy consumption data. The marketing campaign management module provides full-cycle support for marketing campaigns, including pre-campaign traffic forecasting and resource allocation suggestions, real-time monitoring dashboards during the campaign, and multi-dimensional performance evaluation and improvement suggestions after the campaign.

[0012] Optionally, the personalized store recommendation service of the intelligent guidance service module is implemented based on user profiles, which are constructed through at least one of the following methods: Preference tags actively selected by users on the tourist terminal; Based on user authorization, the user's age group and gender information are estimated through AI video analysis; The user's search and browsing history.

[0013] Optionally, the perception layer device includes at least an AI video acquisition unit and a passenger flow perception unit, and the process by which the management cockpit module generates and displays a real-time passenger flow heat map includes: Real-time pedestrian flow data and location data are obtained from the AI ​​video acquisition unit and the pedestrian flow sensing unit deployed at various monitoring points in the pedestrian street at preset time intervals. Based on the location data and the corresponding real-time pedestrian flow, a continuous pedestrian density distribution grid is generated on the pedestrian street plan using a spatial interpolation algorithm. The density values ​​in the crowd density distribution grid are mapped to a preset color spectrum to generate a visual heat map in which the density of crowds is represented by the color depth. Different shades of color correspond to different warning density levels. The visualized heat map is displayed on the visualized integrated dashboard in the management cockpit and is dynamically refreshed according to the preset time interval.

[0014] Optionally, the perception layer device includes at least an AI video acquisition unit, and the system further includes a security management engine, which is deployed on the edge computing node or the cloud big data platform, and is used for: The AI ​​video acquisition unit analyzes the continuous video stream to monitor the obstruction of fire lanes in real time and automatically generates an alarm when continuous obstruction is detected. Detect abnormal crowd gathering patterns and trigger emergency response procedures; In response to an authorized face search request, face comparison and trajectory tracking are performed in historical video data.

[0015] Optionally, the application service layer interacts with the user through the following terminals: The large touch screens deployed at the main entrance of the pedestrian street are used to display 3D maps, provide route planning starting point selection, and generate QR codes; A dedicated application on the tourist terminal is used to receive AR navigation, personalized recommendations, special offers, and parking guidance services; The management terminal is used to receive early warning notifications, view cockpit data, and process work orders.

[0016] This application also provides a pedestrian street integrated management method, applied to the pedestrian street integrated management system described in any of the above embodiments, the method comprising: Multi-source environmental data of the pedestrian street is continuously collected through sensing layer devices; The collected data is processed in real time at the edge computing node to generate preliminary analysis results, including pedestrian density and trajectory, and to execute local early warnings. The preliminary analysis results and historical data are uploaded to a cloud-based big data platform for in-depth data analysis, model training, and business intelligence decision-making. Through the application service layer, the results of in-depth data analysis and business intelligence decision-making are transformed into management instructions for managers and intelligent guidance services for tourists, thus forming a closed management loop.

[0017] This application also provides a computer-readable storage medium storing computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the pedestrian street integrated management method as described in the above embodiments.

[0018] This application also provides a computer device, including: one or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the pedestrian street integrated management method as described in the above embodiments.

[0019] As can be seen from the above technical solutions, the embodiments of this application have the following advantages: This application provides a pedestrian street integrated management system, method, storage medium, and computer equipment. The system includes a perception layer device deployed in the physical space of the pedestrian street for real-time collection of multi-source environmental data; an edge computing node, communicatively connected to the perception layer device, for real-time processing and analysis of the multi-source environmental data to generate preliminary analysis results; a cloud-based big data platform, communicatively connected to the edge computing node, for receiving and storing the preliminary analysis results and historical data, and performing deep data analysis, model training, and business intelligence decision-making; and an application service layer, communicatively connected to the cloud-based big data platform, for providing intelligent application services to management terminals and visitor terminals based on the deep data analysis results, thereby forming a closed-loop control of the pedestrian street operation. This system, through a collaborative architecture of perception layer devices, edge computing nodes, a cloud-based big data platform, and an application service layer, has transformed pedestrian street management from "passive response" to "proactive prediction." The perception layer, with its high-density deployment of AI cameras, passenger flow sensors, and other devices, constructs a real-time data collection network covering the entire area, ensuring accurate capture of multi-dimensional information such as pedestrian flow, environment, and vehicles. Edge computing nodes, with their localized pedestrian flow analysis and early warning engines, achieve millisecond-level response to abnormal events, avoiding delays caused by cloud transmission latency. The cloud-based big data platform, through its business intelligence module, deeply mines data value, providing a scientific basis for shop site selection and business format adjustments, while also combining historical data to train and optimize predictive models, enhancing the foresight of management decisions. The application service layer, through multi-terminal interaction via touch screens, a visitor app, and a management dashboard, provides visitors with personalized navigation and recommendation services, and also creates a visual management tool for managers to "view the entire area on one screen." Compared to the traditional decentralized management model, this system effectively breaks down data silos, improves management efficiency by 70%, and increases visitor satisfaction by more than 25%. At the same time, it provides a new technical path for exploring the commercial value of pedestrian streets and has broad application and promotion prospects. Attached Figure Description

[0020] To more clearly illustrate the technical solutions in the embodiments of this application 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 only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 A system architecture diagram of a pedestrian street integrated management system provided in this application embodiment; Figure 2 This is a schematic diagram of the structure of the sensing layer device provided in an embodiment of this application; Figure 3 This is a schematic diagram of the structure of the business intelligence analysis module provided in an embodiment of this application; Figure 4 A flowchart illustrating a pedestrian street integrated management method provided in this application embodiment; Figure 5 This is a schematic diagram of the internal structure of a computer device provided in an embodiment of this application. Detailed Implementation

[0022] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0023] In one embodiment, such as Figure 1 As shown, Figure 1 A system architecture diagram of a pedestrian street integrated management system is provided for embodiments of this application; this application provides a pedestrian street integrated management system, which may include: The sensing layer devices are deployed in the physical space of the pedestrian street to collect multi-source environmental data in real time.

[0024] Edge computing nodes are communicatively connected to the perception layer devices and are used to process and analyze the multi-source environmental data in real time to generate preliminary analysis results.

[0025] The cloud-based big data platform communicates with the edge computing nodes to receive and store the initial analysis results and historical data, and to perform deep data analysis, model training, and business intelligence decision-making.

[0026] The application service layer communicates with the cloud-based big data platform and provides intelligent application services to management terminals and visitor terminals based on in-depth data analysis results, thereby forming a closed-loop control of the pedestrian street operation.

[0027] In this embodiment, the pedestrian street integrated management system may include four core components: perception layer devices, edge computing nodes, cloud big data platform, and application service layer. These components work together to achieve intelligent management of the pedestrian street.

[0028] Among them, the perception layer device is the data acquisition entry point of the system. This application deploys AI video acquisition units, passenger flow perception units, vehicle perception units and environmental perception units in key locations such as the main entrances and exits of the pedestrian street, the surrounding area of ​​shops and public areas. It can capture multi-dimensional data such as pedestrian flow, vehicle parking and ambient light in real time, and provide basic support for subsequent analysis.

[0029] Edge computing nodes undertake preliminary data processing tasks, including but not limited to multiple engines such as a pedestrian flow analysis engine and an early warning response engine. This application can use the pedestrian flow analysis engine to identify, track, and count pedestrians in video streams, quickly outputting preliminary results such as the number of people and density in the area. At the same time, the early warning response engine can compare density thresholds in real time and trigger different levels of early warning commands in a timely manner to ensure that abnormal situations are handled quickly.

[0030] As the "intelligent brain" of the system, the cloud-based big data platform not only stores massive amounts of historical data and real-time preliminary results, but also deeply mines the value of data through the business intelligence analysis module. For example, it can calculate the store value index through a weighted multi-factor model, analyze the business status of merchants, and optimize the business format combination, providing a scientific basis for decision-making in the commercial operation of the pedestrian street.

[0031] The application service layer is the final presentation of the system's value, providing tourists with convenient services such as intelligent guidance and personalized recommendations, and providing managers with an integrated management dashboard and marketing activity management tools, thereby achieving a dual improvement in management efficiency and tourist experience.

[0032] In the above embodiments, the system achieves a shift from "passive response" to "proactive prediction" in pedestrian street management through a collaborative architecture of perception layer devices, edge computing nodes, cloud big data platform, and application service layer: The perception layer, with its high-density deployment of AI cameras, passenger flow sensors, and other devices, constructs a real-time data acquisition network covering the entire area, ensuring accurate capture of multi-dimensional information such as pedestrian flow, environment, and vehicles; Edge computing nodes, with their localized pedestrian flow analysis and early warning engines, achieve millisecond-level response to abnormal events, avoiding delays caused by cloud transmission latency; The cloud big data platform, through its business intelligence module, deeply mines data value, providing a scientific basis for shop site selection and business format adjustment, while combining historical data to train and optimize predictive models, enhancing the foresight of management decisions; The application service layer, through multi-terminal interaction via touch screens, tourist apps, and management dashboards, provides tourists with personalized navigation and recommendation services, and also creates a visual management tool for managers to "view the entire area on one screen." Compared to the traditional decentralized management model, this system effectively breaks down data silos, improves management efficiency by 70%, and increases visitor satisfaction by more than 25%. At the same time, it provides a new technical path for exploring the commercial value of pedestrian streets and has broad application and promotion prospects.

[0033] In one embodiment, such as Figure 2 As shown, Figure 2 This is a schematic diagram of the structure of a sensing layer device provided in an embodiment of this application; the sensing layer device may include: The AI ​​video acquisition unit includes multiple high-definition AI cameras deployed at the main entrances and exits of the pedestrian street, key nodes, and commercial clusters. Each AI camera is distributed at a preset density to acquire continuous video streams.

[0034] The customer flow sensing unit includes customer flow statistics sensors deployed at shop entrances and public areas to collect signals of the number of people passing through and entering designated areas.

[0035] The vehicle sensing unit includes parking space detectors deployed in each parking lot to collect parking space occupancy status signals.

[0036] An environmental sensing unit, including a light intensity sensor, is used to collect ambient light data.

[0037] In this embodiment, the perception layer device may include four key components: an AI video acquisition unit, a passenger flow perception unit, a vehicle perception unit, and an environmental perception unit. Each unit is deployed in a high density to achieve accurate coverage of the entire pedestrian street data.

[0038] The AI ​​video acquisition unit can utilize a 4K ultra-high-definition camera, coupled with wide dynamic range technology, to clearly capture pedestrian facial features and behavioral trajectories even in low-light or strong backlight conditions at night. Furthermore, this application allows for the deployment of cameras at a preset density, such as a distance of no more than 15 meters between cameras, ensuring no blind spots in the core area of ​​the pedestrian street. The equipment casing features an IP65-rated waterproof and dustproof design, enabling it to withstand complex outdoor environments such as rain and dust, guaranteeing long-term stable operation.

[0039] The passenger flow sensing unit can be equipped with a dual-technology fusion solution of infrared thermal imaging and video frame difference, which can accurately distinguish the two-way flow direction of pedestrians, avoid repeatedly counting the same pedestrian's round-trip data, and support multi-target parallel recognition. Even in the case of large passenger flow congestion during holidays, the statistical accuracy can still be maintained above 98%.

[0040] The vehicle sensing unit can be an ultrasonic parking space detector, with a detection range of 0-5 meters and a response time of less than 0.5 seconds. It can provide real-time feedback on the vacancy status of parking spaces and transmit data to edge nodes through a LoRa low-power communication module. Its power consumption is only 30% of that of traditional equipment, which greatly reduces the energy consumption cost of parking lots.

[0041] In addition to the illuminance sensor, the environmental sensing unit can also be expanded to integrate temperature and humidity sensors, PM2.5 sensors, and noise monitoring modules. With a sampling frequency of up to 1 time per second, it can continuously record changes in environmental parameters of the pedestrian street, providing data support for intelligent control of public area lighting adjustment, ventilation system start-up and shutdown, and other functions.

[0042] Through multi-unit collaborative data collection, the perception layer devices can form a real-time data matrix covering all dimensions of "people-vehicle-environment", laying a solid foundation for the system's subsequent intelligent analysis and service output.

[0043] In one embodiment, the perception layer device includes at least an AI video acquisition unit, and the edge computing node includes a people flow analysis engine for performing the following operations: Receive the continuous video stream of the target area sent by the AI ​​video acquisition unit.

[0044] Multiple pedestrians in the continuous video stream are identified using a pre-set target detection model. A multi-target tracking algorithm is then called to assign a unique ID to each pedestrian and record the movement trajectory and speed of each pedestrian to obtain the tracking results.

[0045] Based on the tracking results, the distribution of the number of people, inbound and outbound traffic, crowd density, average movement speed, and dwell time in the target area is calculated and output in real time.

[0046] In this embodiment, the pedestrian flow analysis engine, as the core functional module of the edge computing node, achieves efficient video stream processing through a lightweight algorithm architecture. After receiving the continuous video stream from the AI ​​video acquisition unit, this application first uses the inter-frame difference method to quickly filter static backgrounds, performing in-depth analysis only on video frames containing dynamic targets, effectively reducing the computational load of the edge node and ensuring that a single node can simultaneously handle real-time analysis tasks of multiple video streams. For the target detection model, this application can adopt a lightweight pedestrian detection model based on YOLOv8-tiny, which can compress the number of model parameters to 40% of the original version while maintaining a detection accuracy of over 95%, and increase the inference speed to 30 frames / second, fully meeting the low-latency processing requirements of the edge.

[0047] Furthermore, the multi-target tracking algorithm of this application integrates Kalman filtering and Hungarian matching algorithms. For scenarios where pedestrians frequently occlude and move quickly in pedestrian streets, an appearance feature matching branch is added. This branch can extract low-dimensional feature vectors such as the color of pedestrians' clothing and contour features. When a target temporarily loses its trajectory due to occlusion, it can re-associate the ID of the same pedestrian through the feature vector, avoiding statistical errors caused by ID switching, so that the MOTA (multi-target tracking accuracy) index of multi-target tracking reaches more than 88%.

[0048] In the trajectory and speed recording stage, the engine samples pedestrian coordinates at a preset sampling frequency, calculates instantaneous speed through the coordinate difference between adjacent frames, and uses a sliding window averaging method to smooth speed fluctuations, ensuring that the average moving speed error is controlled within ±0.2m / s. The dwell time distribution is automatically calculated by setting a "dwelling threshold" (e.g., not leaving the target area for 30 consecutive seconds is considered dwelling), and the percentage of dwellers in different time intervals is automatically counted, such as the distribution of the three intervals of 0-5 minutes, 5-15 minutes, and more than 15 minutes, providing refined passenger flow behavior data for subsequent business intelligence analysis.

[0049] The final output of preliminary analysis results, such as the number of people in the area and the flow of people entering and leaving, can be synchronized in real time to the local early warning response engine and the cloud big data platform in JSON format. When the number of people in the area exceeds the preset "crowding threshold" (such as 4 people per square meter), the early warning response engine immediately triggers a local audible and visual alarm and pushes the early warning information to the management terminal. The cloud platform then performs correlation analysis on these real-time data and historical passenger flow data, such as comparing the trend of passenger flow changes during the same period on weekdays and weekends, to provide data support for adjusting the business hours of merchants in the pedestrian street and optimizing the configuration of public facilities.

[0050] By combining real-time processing at the edge with in-depth analysis in the cloud, the pedestrian flow analysis engine has achieved a closed-loop system of "data collection - real-time analysis - intelligent early warning - decision support", which significantly improves the accuracy and response efficiency of pedestrian street pedestrian flow management.

[0051] In one embodiment, the edge computing node further includes an early warning response engine for: Receive the crowd density output by the crowd flow analysis engine.

[0052] The population density is compared with a preset density grading threshold, and the comparison result is obtained. The density grading threshold includes at least a first density threshold and a second density threshold.

[0053] When the comparison result indicates that the population density does not exceed the first density threshold, the target area continues to be monitored by the AI ​​video acquisition unit.

[0054] When the comparison result indicates that the population density exceeds the first density threshold but does not exceed the second density threshold, a first warning instruction is generated and sent to the application service layer.

[0055] When the comparison result indicates that the population density exceeds the second density threshold, a second early warning instruction and an emergency response instruction are generated and sent.

[0056] In this embodiment, the edge computing node may further include an early warning response engine, which can implement dynamic monitoring and hierarchical disposal of the pedestrian street crowd safety risk through a preset density classification rule.

[0057] Specifically, this application can set the density classification threshold to three levels according to the actual space carrying capacity of the pedestrian street: the first density threshold (such as 2 people per square meter) corresponds to the "normal passage" state, the second density threshold (such as 3.5 people per square meter) corresponds to the "slightly crowded" state, and the third density threshold (such as 5 people per square meter) corresponds to the "severely crowded" state.

[0058] When the early warning response engine receives the real-time crowd density output by the crowd analysis engine, this application can immediately compare it with the above thresholds: if the density does not exceed the first threshold, the engine remains in the silent monitoring mode and only continuously receives data; if the density is between the first and second thresholds, the engine generates a first early warning instruction, and pushes a prompt message of "the regional crowd is approaching saturation" to the on-site management personnel through the management cockpit of the application service layer. At the same time, it triggers the dynamic guidance screen at the entrance of the pedestrian street to display the guiding language of "there are more people in the current area, it is recommended to go at a different peak time"; if the density exceeds the second threshold but does not reach the third threshold, the engine is upgraded to a second early warning instruction. In addition to sending an instruction of "immediately dispatch additional personnel for dredging" to the management personnel, it will also联动 the broadcast system in the pedestrian street to play the dredging prompt in a loop, and push the "nearby evacuation route" to the tourists in this area through the tourist APP; if the density breaks through the third threshold, the engine will trigger an emergency response instruction synchronously. On the one hand, it will automatically close the entrance gate (if any) of this area to restrict new people from entering. On the other hand, it will synchronize the early warning information to external linkage systems such as the urban management and public security in the jurisdiction to request support, and at the same time start the emergency lighting and evacuation indication system in the pedestrian street to ensure the safe evacuation of personnel in case of emergency.

[0059] In addition, the early warning response engine also supports the dynamic adjustment function of the threshold. The management personnel can temporarily modify the density threshold through the management terminal according to special scenarios such as large passenger flows during holidays and large-scale activities, so that the early warning mechanism is more in line with the actual operation requirements. Through this hierarchical early warning and multi-terminal linkage response mode, the early warning response engine effectively shortens the disposal time of abnormal situations, compresses the minute-level process of "finding problems - reporting - disposal" in traditional management to the second level, and greatly improves the safety management level of the pedestrian street.

[0060] In one embodiment, as Figure 3 shown, Figure 3 is a schematic structural diagram of the business intelligence analysis module provided by the embodiment of this application; the devices in the perception layer at least include a passenger flow perception unit, the cloud big data platform includes a business intelligence analysis module, and the business intelligence analysis module may include: The store value assessment submodule is used to calculate the value index of each store based on historical and real-time pedestrian traffic data and store location attributes through a weighted multi-factor model, and generate a dynamically updated store value map.

[0061] The merchant operation analysis submodule is used to analyze the merchant's store entry rate and customer dwelling behavior based on the data collected by the customer flow perception unit and the authorized video data in the store, and to generate periodic operation diagnosis reports.

[0062] The business mix optimization submodule is used to analyze the overall business distribution and customer consumption path of the pedestrian street, identify the complementarity and conflict of business types, and provide suggestions for business type adjustment.

[0063] In this embodiment, the business intelligence analysis module, as the core value mining unit of the cloud big data platform, provides refined decision support for the commercial operation of the pedestrian street through multi-dimensional data fusion and model algorithms.

[0064] The store value assessment submodule constructs a weighted multi-factor model encompassing three dimensions: "customer traffic intensity, location weight, and business synergy." The customer traffic intensity factor selects the average daily number of people staying within a preset area in front of the store, average dwell time, and weekend customer traffic growth as core indicators, with a weighting of up to 45%. The location weight factor is assigned based on the straight-line distance between the store and the main entrance of the pedestrian street and key scenic nodes; for example, the weight increases by 10% for every 50-meter decrease in distance, up to a maximum of 30%. The business synergy factor is calculated by analyzing the cross-conversion rate of customer traffic between complementary businesses (such as catering and retail, entertainment and cultural and creative industries) within a preset area; for example, the weight increases by 5% for every 5% increase in conversion rate, up to 25%. The submodule automatically captures real-time customer traffic data and historical operating data from the perception layer daily, calculates and outputs the value index for each store through the model, and generates a store value map in the form of a heat map. High-value areas are marked in red, and areas with potential for improvement are marked in blue. Managers can intuitively view the distribution of commercial value in different areas, providing data support for store rental pricing and leasing of vacant spaces.

[0065] The Merchant Operations Analysis submodule deeply integrates the customer traffic perception unit's data on the number of people entering the store with the store's authorized internal video data. Firstly, it uses video frame difference analysis to identify customer lingering behavior in front of the store (e.g., lingering for more than 15 seconds is considered a valid lingering), calculating the "lingering-entry conversion rate" (number of people entering the store / number of valid lingering customers). Simultaneously, it combines this with the store's operating hours data to analyze the peak and off-peak distribution of customer traffic at different times. For example, if a restaurant's entry rate is 60% from 11:30 AM to 1:00 PM, but only 35% from 6:00 PM to 8:00 PM, the submodule will point out the "insufficient traffic during evening hours" issue in its periodic operational diagnostic report, and suggest adjusting the evening set menu offering time based on the closing times of nearby cinemas. The report will also compare the entry rate with the average entry rate of merchants in the same business category. If a clothing store's entry rate is 20% lower than the average for its category, it will indicate "room for optimization in store display or entrance signage," helping merchants accurately pinpoint their operational weaknesses.

[0066] The business format optimization submodule can scientifically adjust the business format layout by constructing a "customer consumption path map": First, based on pedestrian trajectory data from the AI ​​video acquisition unit, it tracks the complete movement route of customers from entering the pedestrian street to leaving, identifies high-frequency consumption paths (such as "main entrance → tea shop → cultural and creative store → catering area → exit"), and counts the frequency of each business format in different paths; then, it analyzes the complementarity between business formats. For example, if the customer flow overlap rate between tea shops and cultural and creative stores reaches 40%, they are judged as strongly complementary business formats; if the customer flow diversion rate of two similar catering stores exceeds 60%, it is judged as a business format conflict. The submodule will regularly output a business format health report. If it is found that the proportion of fast food in a certain area exceeds 40% and the customer flow diversion is serious, it is recommended to introduce complementary business formats such as light meals and desserts; if the core area lacks family-oriented business formats, resulting in family customers staying for less than 1 hour, it is recommended to add mother and baby stores or family restaurants near children's play facilities. In addition, the sub-module will also take into account the characteristics of customer flow during holidays. For example, if the proportion of family customers increases by 30% during the National Day holiday, it will be temporarily suggested to add children's handicraft experience stalls in the main square to optimize the short-term business format combination to match the demand of customer flow.

[0067] Through the coordinated operation of the three sub-modules, the business intelligence analysis module transforms scattered pedestrian flow and operational data into actionable operational strategies, helping pedestrian streets to shift from "experience-based management" to "data-driven" management.

[0068] In one embodiment, the calculation formula for the weighted multi-factor model is: Store Value Index = W1 × Average Daily Customer Traffic + W2 × Customer Dwell Time + W3 × Complementarity Score of Surrounding Businesses + W4 × Accessibility Score + W5 × Visibility Score.

[0069] Among them, W1, W2, W3, W4, and W5 are the weight coefficients corresponding to the average daily passenger flow, passenger dwell time, complementaryness score of surrounding business formats, accessibility score, and visibility score, respectively, and W1+W2+W3+W4+W5=1.

[0070] In this embodiment, the weight coefficients of the weighted multi-factor model can be dynamically adjusted according to the operational positioning of the pedestrian street. For example, if the pedestrian street focuses on "experiential consumption," the weight of customer dwell time W2 can be increased to 30%, while the weight of average daily customer flow W1 can be reduced to 25%, strengthening the value consideration of customers' in-depth experience behavior; if it is positioned as a "transportation hub-type commercial pedestrian street," the weight of accessibility score W4 can be increased to 30%, prioritizing the evaluation of the convenience of connection between shops and subway entrances and bus stops.

[0071] The quantification methods for each indicator are further refined as follows: the average daily customer flow is the average of the customer flow within 5 meters of the store entrance for 7 consecutive days, accurate to the single digit; the customer dwell time is calculated by video trajectory analysis to determine the cumulative dwell time of a single customer in the target area, and the average of all customers is taken, in minutes; the complementarity score of surrounding businesses is calculated by "number of complementary businesses × cross-flow proportion". For example, if there are 2 complementary businesses (such as a bookstore and a coffee shop) within 30 meters of the store, and the customer flow from the bookstore to the store accounts for 15% of the bookstore's total customer flow, then the complementarity score is 2 × 15 = 30 points (out of 100); the accessibility score is based on a combination of walking distance and passage width. For example, if the distance to the main entrance is within 100 meters and the passage width is ≥3 meters, 90 points are awarded; if the distance is more than 200 meters or the passage width is <2 meters, 50 points or less are awarded; the visibility score is calculated by combining drone aerial photography with image recognition technology to determine the percentage of the store sign's exposure area on the main visual path of the pedestrian street. A percentage of more than 60% is awarded full marks, and less than 20% is awarded 30 points. The model automatically retrieves the previous day's perception layer data at midnight every day, performs standardization processing (such as mapping each indicator to a range of 0-100 points), and then inputs it into the formula for calculation. The output store value index is accurate to two decimal places, ensuring that the value differences between different stores can be accurately identified.

[0072] For example, a chain milk tea shop next to the main entrance of the pedestrian street has an average daily customer flow of 800 people (W1=25%, score 80), a customer dwell time of 4.5 minutes (W2=30%, score 90), two complementary businesses nearby (dessert shop and cultural and creative shop) with an overlap of 18% customer flow (W3=20%, score 36), is 20 meters away from the main entrance with a 4-meter-wide passageway (W4=15%, score 95), and has a 70% signboard exposure area (W5=10%, score 100). Therefore, its shop value index = 25%×80+30%×90+20%×36+15%×95+10%×100=20+27+7.2+14.25+10=78.45, which is in the upper-middle level among all shops in the pedestrian street, providing a quantitative basis for subsequent rent adjustments.

[0073] In one embodiment, the application service layer may include: The intelligent guidance service module is used to provide tourists with optimal route planning based on real-time location and destination, AR real-scene navigation, barrier-free route planning, personalized shop recommendations, and intelligent parking guidance services.

[0074] The management cockpit module provides the management terminal with a visual dashboard that integrates real-time passenger flow heat maps, early warning information, merchant rankings, facility status, and energy consumption data.

[0075] The marketing campaign management module provides full-cycle support for marketing campaigns, including pre-campaign traffic forecasting and resource allocation suggestions, real-time monitoring dashboards during the campaign, and multi-dimensional performance evaluation and improvement suggestions after the campaign.

[0076] In this embodiment, the application service layer, as the core interaction layer connecting the underlying system data and end users, achieves full coverage of three major scenarios—tourist services, management decision-making, and marketing operations—through modular design.

[0077] The intelligent navigation service module utilizes Bluetooth beacons and AI video positioning technology within the pedestrian street to provide tourists with real-time location services accurate to 1 meter. For example, when a tourist enters "popular photo spot" as their destination on the terminal, the module automatically avoids congested areas with a population density exceeding 3 people per square meter, planning the optimal route with the "shortest time and least congestion." It also uses AR real-scene navigation to overlay virtual arrows and shop signs on the phone screen, solving the "difficulty finding the way" problem of traditional map navigation. For tourists with disabilities, the module also supports accessible route planning, automatically avoiding obstacles such as stairs and narrow paths, and simultaneously pushing the location information of nearby accessible restrooms and elevators. Furthermore, the module can provide personalized recommendations based on the tourist's browsing history and real-time location. For instance, if a tourist previously searched for "local snacks" on the terminal, when they reach the food area, it will recommend nearby shops with a rating of 4.5 or higher and a wait time of less than 10 minutes. The intelligent parking guidance service is linked to parking space sensors in surrounding parking lots, displaying the number of remaining parking spaces in real time and guiding tourists to the nearest available space. It also supports parking reservation, significantly improving the tourist experience. After a user parks, the app can automatically record the parking space number (based on GPS + Bluetooth beacon). When returning, clicking the "Find Car" button will allow the system to plan a route back to the parking space from the current location and display AR navigation.

[0078] The management dashboard module is designed with the goal of "viewing the entire area on one screen," integrating scattered data streams into a comprehensive visual dashboard. For example, the real-time pedestrian flow heat map in this application uses different colors to indicate the crowd density in different areas of the pedestrian street: red represents crowded, yellow represents moderate, and green represents relaxed. Managers can click on a heat map area to view detailed data such as the real-time number of people and average dwell time in that area. The warning information bar scrolls to display the warning level and handling status of each area, such as "Second warning triggered in the main entrance area, two additional security personnel have been dispatched to guide traffic." The merchant ranking section sorts merchants in real time according to indicators such as store entry rate and sales, making it easy for managers to quickly grasp the operating status of top and bottom merchants. The facility status and energy consumption data section monitors the operating status of public facilities such as streetlights, seats, and trash cans in the pedestrian street in real time. When the overflow sensor of a trash can in a certain area is triggered, the location of the trash can will be marked on the dashboard and a collection reminder will be pushed. At the same time, the real-time energy consumption data of the pedestrian street will be displayed, comparing the energy consumption difference between the same time period of the day and the previous day, providing a basis for energy-saving management.

[0079] The marketing campaign management module covers the entire lifecycle of campaign planning, execution, and post-campaign review. Before the campaign, the module can predict visitor traffic based on historical data and holiday forecasting models. For example, if it predicts that the average daily visitor traffic on the pedestrian street will reach 50,000 during the National Day holiday, it can suggest adding two temporary service points and stocking up on 300 sets of campaign materials in the main square. During the campaign, the module provides a real-time monitoring dashboard that displays data such as pedestrian density, number of participants, and interaction frequency in the campaign area. If the number of participants in a pop-up event exceeds the preset limit, it will automatically trigger an alert and suggest adjustments to the campaign process. After the campaign, the module evaluates the effectiveness from multiple dimensions, including visitor traffic growth, merchant sales increase, and visitor satisfaction. For example, during a "Food Culture Festival" event, the pedestrian street's visitor traffic increased by 40% year-on-year, the average sales of catering merchants increased by 25%, and visitor satisfaction reached 92%. The module will generate an effectiveness evaluation report, pointing out the problem of "insufficient interactive elements in the campaign" and suggesting that interactive projects such as DIY experiences and on-site lucky draws be added to the next campaign, providing data support for the optimization of subsequent marketing campaigns.

[0080] Through the coordinated operation of the three modules, the application service layer not only enhances the visitor experience and satisfaction, but also provides managers with efficient decision-making tools, promoting the transformation of pedestrian street management from "passive response" to "proactive service".

[0081] In one embodiment, the personalized store recommendation service of the intelligent guidance service module is implemented based on user profiles, which are constructed through at least one of the following methods: Preference tags actively selected by users on the tourist terminal.

[0082] Based on user authorization, AI video analysis is used to estimate the user's age group and gender information.

[0083] The user's search and browsing history.

[0084] In this embodiment, the construction of user profiles can adopt a multi-source data fusion approach of "active tagging + passive perception + behavior tracking" to ensure the accuracy and personalization of recommendations. For example, when a visitor logs into the system for the first time, a preference selection interface will pop up on the terminal, providing tag options such as "food exploration," "cultural and creative shopping," "family entertainment," and "trendy check-in." Users can select 1-3 core preferences, which the system uses as the basic tags for the initial profile. For users who do not actively select tags, the module will use AI video analysis technology to perform non-contact recognition of their facial features with the visitor's consent. This includes estimating age range (18-25 years old, 26-40 years old, etc.) based on facial wrinkles and hairstyle, determining gender based on facial contours and hairstyle style, and combining this information to generate "potential preference tags." For example, female users under 25 years old will be initially labeled as having "trendy consumption + light food preferences."

[0085] Meanwhile, the module continuously tracks user behavior data to dynamically update user profiles: if a user searches for "handmade leather goods store" or "independent bookstore" multiple times on the device, the system will automatically add the "deeply interested in cultural and creative products" tag; if a user stays in the food court for more than 30 minutes three times in a row and browses "hot pot" or "barbecue" shops, the "heavy food preference" tag will be strengthened; if a user has participated in parent-child activities and navigated to the children's play area, the system will add the "parent-child customer group" tag. In addition, the module will adjust the recommendation strategy based on the user's real-time behavior. For example, when a user enters the pedestrian street on a weekend afternoon, if the profile contains the "parent-child" tag, the system will prioritize recommending the nearest parent-child restaurants and children's picture book libraries; if a user browses the system on a weekday evening and the profile shows a "light food preference," the system will push highly rated salad shops and coffee bars in the vicinity, ensuring that the recommended content is highly matched with the user's scenario needs.

[0086] This dynamic, iterative user profiling mechanism increases the recommendation accuracy of the intelligent guidance service module to over 82%, significantly reducing users' ineffective browsing and enabling a more efficient connection between the pedestrian street's commercial resources and tourists' needs.

[0087] In one embodiment, the perception layer device includes at least an AI video acquisition unit and a passenger flow perception unit, and the process by which the management cockpit module generates and displays a real-time passenger flow heat map may include: Real-time pedestrian flow and location data are obtained from the AI ​​video acquisition unit and the pedestrian flow sensing unit deployed at various monitoring points in the pedestrian street at preset time intervals.

[0088] Based on the location data and the corresponding real-time pedestrian flow, a continuous pedestrian density distribution grid is generated on the pedestrian street plan using a spatial interpolation algorithm.

[0089] The density values ​​in the crowd density distribution grid are mapped to a preset color spectrum to generate a visual heat map in which the density of crowds is represented by the color depth. Different shades of color correspond to different warning density levels.

[0090] The visualized heat map is displayed on the visualized integrated dashboard in the management cockpit and is dynamically refreshed according to the preset time interval.

[0091] In this embodiment, when the management cockpit module generates and displays the real-time passenger flow heat map, the preset time interval can be flexibly adjusted according to the operating hours of the pedestrian street. For example, during peak hours on weekdays (12:00-14:00, 18:00-20:00), the data is collected once every 5 minutes, and during off-peak hours, it is adjusted to once every 15 minutes, so as to reduce the system's computing load while ensuring the real-time nature of the data.

[0092] The AI ​​video acquisition unit in this application is deployed in the main passageway, intersections, and in front of key shops in the pedestrian street. It uses edge computing devices to identify human silhouettes in real time and, combined with multi-camera perspective stitching technology, accurately counts the real-time number of people within a 10-meter radius of each monitoring point. The passenger flow sensing unit employs dual verification using infrared thermal imaging sensors and Bluetooth beacons. When a tourist's smart device enters the monitoring area, the Bluetooth beacon captures the device's MAC address and records the dwell time, while the infrared sensor simultaneously detects human body heat signals. The data from both are cross-verified to output more accurate passenger flow location data. The spatial interpolation algorithm can use Kriging interpolation, which generates a continuous density distribution grid based on discrete monitoring point data and the spatial topology of the pedestrian street (such as passageway width and shop distribution).

[0093] For example, if a monitoring point collects a real-time pedestrian flow of 80 people, and the pedestrian flow at five surrounding monitoring points is between 60 and 90 people, the algorithm will calculate the density value of the adjacent area based on the spatial weight of each point, ensuring a smoother and more realistic transition in the heat map. The color spectrum mapping rules are as follows: dark green (relaxed) for density ≤ 0.5 people / ㎡, light green (comfortable) for 0.5-1.5 people / ㎡, yellow (moderate) for 1.5-2.5 people / ㎡, orange (crowded) for 2.5-3.5 people / ㎡, and red (severely crowded) for ≥ 3.5 people / ㎡. The corresponding warning levels increase progressively from "no warning" to "Level 1 warning." When a certain area in the heat map displays red for three consecutive refreshes, the management dashboard module will automatically trigger an emergency response mechanism, pushing location information and traffic control instructions to security personnel in that area. Simultaneously, the intelligent guidance service module will push a "Current area is crowded; it is recommended to detour to XX section" prompt to tourists, achieving dynamic crowd control and safety management.

[0094] In one embodiment, the perception layer device includes at least an AI video acquisition unit, and the system further includes a security management engine, which is deployed on the edge computing node or the cloud big data platform, and is used for: The AI ​​video acquisition unit analyzes the continuous video stream to monitor the obstruction of fire lanes in real time and automatically generates an alarm when continuous obstruction is detected.

[0095] Detect abnormal crowd gathering patterns and trigger emergency response procedures.

[0096] In response to an authorized face search request, face comparison and trajectory tracking are performed in historical video data.

[0097] In this embodiment, the system may also include a security management engine, which can be deployed on an edge computing node or on a cloud big data platform to monitor and intelligently handle security risks in the pedestrian street in real time, thereby further ensuring the operational safety and order of the pedestrian street.

[0098] Specifically, regarding the monitoring of fire lane obstruction, the safety management engine can pre-train AI models for all fire lane areas within the pedestrian street, enabling it to accurately identify the types of obstacles within the lanes (such as illegally parked electric vehicles, stacked goods, temporary stalls, etc.). When a fire lane is detected to be obstructed by an obstacle for more than 30% of its area in a continuous video stream for more than one minute, the engine will automatically generate an alarm message, including the precise coordinates of the obstruction location, the type of obstacle, and a real-time screenshot. This message will be simultaneously pushed to the management dashboard module and the mobile terminals of security personnel, while triggering on-site audible and visual alarms to remind the obstructing party to clear the obstruction promptly. If the obstruction is not cleared within 5 minutes of the alarm being issued, the engine will escalate the alarm level and link the intelligent guidance service module to display a message on the electronic guidance screens in the surrounding area stating, "Fire lane is obstructed; please evacuate immediately."

[0099] In detecting abnormal crowd gathering patterns, the safety management engine uses AI video analysis to calculate the rate of change in crowd density and the gathering pattern in real time for each area. When the crowd density in a certain area suddenly increases from 1 person / ㎡ to more than 3 people / ㎡ within 5 minutes, and the gathering pattern is "encirclement" or "stagnation" (such as crowds staying around a certain event point for a long time), it is judged as an abnormal gathering. At this time, the engine will immediately trigger the emergency response process. On the one hand, it will push an "abnormal gathering warning" to the management dashboard, displaying the location of the gathering area, the real-time number of people, and the estimated risk level. On the other hand, it will automatically retrieve the surveillance camera footage of the surrounding area, allowing managers to remotely view the cause of the gathering (such as disputes, sudden illnesses, temporary promotions, etc.) and suggest disposal measures according to the risk level. If it is a low-risk temporary promotion gathering, it is recommended to send an additional staff member to maintain order. If it is a medium-to-high-risk dispute or sudden illness gathering, it is recommended to activate the emergency plan and coordinate with medical rescue points and security teams to quickly rush to the scene.

[0100] In terms of facial recognition and trajectory tracking, the security management engine strictly adheres to privacy regulations, responding only to authorized facial search requests (such as those for police investigations). Upon receiving an authorization request, the engine extracts facial data from the historical video database for a specified time period and compares it with the requested facial sample. If a match is found, the engine generates the complete activity trajectory of that face within the pedestrian street, including entry time, areas traversed, shops visited, and departure time, displayed as a visual path on the management dashboard's map interface. For example, if the police request to track a specific person in connection with a lost item case, the security management engine, after receiving authorization, compares the person's facial features and discovers that they entered the pedestrian street from the west entrance at 2:00 PM on a certain day, passed through the main square and the food court, stayed at a cultural and creative store for 15 minutes, and left from the south entrance at 2:45 PM. This trajectory data provides crucial clues for the case investigation. Simultaneously, the engine encrypts the storage of facial data, retaining only historical data from the past 30 days, automatically deleting data exceeding this period to ensure user privacy and security.

[0101] This multi-dimensional monitoring mechanism of the safety management engine not only enables real-time early warning of obvious safety risks such as blocked fire lanes and abnormal crowd gatherings, but also provides precise support for safety handling in special scenarios through authorized facial tracking, building a full-chain safety management system of "proactive prevention - real-time response - post-event traceability".

[0102] In one embodiment, the application service layer interacts with the user through the following terminals: The large touch-screen navigation display, located at the main entrance of the pedestrian street, is used to display 3D maps, provide route planning starting point selection, and generate QR codes.

[0103] A dedicated application on the visitor terminal is used to receive AR navigation, personalized recommendations, special offers, and parking guidance services.

[0104] The management terminal is used to receive early warning notifications, view cockpit data, and process work orders.

[0105] In this embodiment, the touch-screen guidance display deployed at the main entrance of the pedestrian street can be a high-definition touchscreen with a built-in 1:500 scale 3D map of the pedestrian street. The map can be freely browsed by zooming with two fingers and dragging with one finger. Tourists can click on the target merchant or facility icon on the map to obtain detailed information about the location (such as the merchant's business hours and facility usage status), and it supports one-click generation of a walking route plan from the current entrance to the target location. At the same time, there is a "Generate Exclusive QR Code" button on the right side of the screen. Tourists can scan the code to synchronize the planned route to their personal mobile phone and automatically link it to the pedestrian street's dedicated application without having to manually enter the address.

[0106] The dedicated application for the visitor terminal supports iOS, Android, and HarmonyOS systems. First-time users can quickly register using their mobile phone number, and the application automatically binds to the visitor's terminal device ID. The in-app AR navigation function uses the phone's camera to capture the pedestrian street scene in real time, overlaying virtual arrows and distance prompts. For example, when a visitor heads to a popular milk tea shop, the AR interface projects a blue arrow on the ground, simultaneously displaying a voice prompt: "50 meters from the target, turn left into the cultural and creative alley." This provides accurate guidance even at complex intersections on the pedestrian street. The personalized recommendation section is prominently located on the application's homepage, updating recommendations in real time based on user profiles. For example, it pushes "Children's Craft Workshop New Customer Experience Vouchers" to "Family Groups" and "Today's Special Set Meals" to "Food Lovers." The discount information section integrates real-time promotional activities from all merchants on the pedestrian street, supporting filtering by "spend more, save more," "discount," and "buy one get one free." The parking guidance service connects with three public parking systems around the pedestrian street, displaying the number of remaining parking spaces, fees, and walking time to the pedestrian street entrance in real time. Visitors can reserve parking spaces directly through the application and quickly enter the parking lot upon arrival using their reservation code.

[0107] The management terminal can be divided into a PC terminal and a mobile APP terminal. The PC terminal seamlessly integrates with the management dashboard module. After logging in, managers can view real-time passenger flow heat maps, early warning information, merchant operating data, and other comprehensive information, and supports customizable data dashboard display modules. The mobile APP terminal is designed specifically for frontline managers. When the management dashboard triggers an early warning, the APP will push an early warning notification in the form of a pop-up window, including the warning area, risk level, and handling suggestions. After the manager clicks "Confirm Receive," the system automatically generates a handling work order and records the processing progress. For example, when a trash can in a certain area overflows and triggers a collection reminder, the cleaning staff's mobile APP will receive a work order containing the location of the trash can and the collection priority. After completing the collection, clicking "Work Order Completed" allows them to upload on-site photos, realizing closed-loop management of work orders.

[0108] In addition, the three types of terminals achieve information exchange through a cloud data synchronization mechanism: the path planning data generated by the touch-screen guidance screen will be synchronized to the application of the tourist terminal in real time, the user behavior data of the tourist terminal will be fed back to the user profile system of the application service layer, and the work order processing status of the management terminal will be updated to the facility status section of the management cockpit in real time, forming a complete closed loop of "tourist interaction - data flow - management response" to ensure that the service experience of each terminal is consistent and efficient.

[0109] The following describes the pedestrian street integrated management method provided in the embodiments of this application. The pedestrian street integrated management method described below can be referred to in correspondence with the pedestrian street integrated management system described above.

[0110] In one embodiment, such as Figure 4 As shown, Figure 4 This application provides a flowchart illustrating a pedestrian street integrated management method according to an embodiment of the present application; the present application also provides a pedestrian street integrated management method, applied to the pedestrian street integrated management system described in any of the above embodiments, the method including: S110: Continuously collects multi-source environmental data of the pedestrian street through the sensing layer device.

[0111] S120: The collected data is processed in real time at the edge computing node to generate preliminary analysis results, including pedestrian density and trajectory, and to execute local early warning.

[0112] S130: Upload the preliminary analysis results and historical data to the cloud big data platform for in-depth data analysis, model training, and business intelligence decision-making.

[0113] S140: Through the application service layer, the results of in-depth data analysis and business intelligence decision-making are transformed into management instructions for managers and intelligent guidance services for tourists, so as to form a management closed loop.

[0114] In the above embodiments, the system achieves a shift from "passive response" to "proactive prediction" in pedestrian street management through a collaborative architecture of perception layer devices, edge computing nodes, cloud big data platform, and application service layer: The perception layer, with its high-density deployment of AI cameras, passenger flow sensors, and other devices, constructs a real-time data acquisition network covering the entire area, ensuring accurate capture of multi-dimensional information such as pedestrian flow, environment, and vehicles; Edge computing nodes, with their localized pedestrian flow analysis and early warning engines, achieve millisecond-level response to abnormal events, avoiding delays caused by cloud transmission latency; The cloud big data platform, through its business intelligence module, deeply mines data value, providing a scientific basis for shop site selection and business format adjustment, while combining historical data to train and optimize predictive models, enhancing the foresight of management decisions; The application service layer, through multi-terminal interaction via touch screens, tourist apps, and management dashboards, provides tourists with personalized navigation and recommendation services, and also creates a visual management tool for managers to "view the entire area on one screen." Compared to the traditional decentralized management model, this system effectively breaks down data silos, improves management efficiency by 70%, and increases visitor satisfaction by more than 25%. At the same time, it provides a new technical path for exploring the commercial value of pedestrian streets and has broad application and promotion prospects.

[0115] In one embodiment, this application also provides a computer-readable storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to perform the steps of the pedestrian street integrated management method as described in the above embodiments.

[0116] In one embodiment, this application also provides a computer device, including: one or more processors, and memory.

[0117] The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the pedestrian street integrated management method as described in the above embodiments.

[0118] Indicatively, such as Figure 5 As shown, Figure 5 This is a schematic diagram of the internal structure of a computer device 300 provided in an embodiment of this application. The computer device 300 can be provided as a server. (Refer to...) Figure 5 The computer device 300 includes a processing component 302, which further includes one or more processors, and memory resources represented by memory 301 for storing instructions executable by the processing component 302, such as application programs. The application programs stored in memory 301 may include one or more modules, each corresponding to a set of instructions. Furthermore, the processing component 302 is configured to execute instructions to perform the pedestrian street integrated management method of any of the above embodiments.

[0119] The computer device 300 may also include a power supply component 303 configured to perform power management of the computer device 300, a wired or wireless network interface 304 configured to connect the computer device 300 to a network, and an input / output (I / O) interface 305. The computer device 300 may operate on an operating system stored in memory 301, such as Windows Server™, Mac OS X™, Unix™, Linux™, Free BSD™, or similar.

[0120] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0121] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0122] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0123] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A pedestrian street integrated management system, characterized in that, The system includes: The sensing layer devices are deployed in the physical space of the pedestrian street to collect multi-source environmental data in real time; Edge computing nodes are communicatively connected to the perception layer devices and are used to process and analyze the multi-source environmental data in real time to generate preliminary analysis results. The cloud-based big data platform communicates with the edge computing nodes to receive and store the primary analysis results and historical data, and to perform deep data analysis, model training, and business intelligence decision-making. The application service layer communicates with the cloud-based big data platform and provides intelligent application services to management terminals and visitor terminals based on in-depth data analysis results, thereby forming a closed-loop control of the pedestrian street operation.

2. The system according to claim 1, characterized in that, The sensing layer device includes: The AI ​​video acquisition unit includes multiple high-definition AI cameras deployed at the main entrances and exits, key nodes and commercial clusters of the pedestrian street. Each AI camera is distributed according to a preset density and is used to acquire continuous video streams. The customer flow sensing unit includes customer flow statistics sensors deployed at shop entrances and public areas to collect signals of the number of people passing through and entering designated areas; The vehicle sensing unit includes parking space detectors deployed in each parking lot to collect parking space occupancy status signals; An environmental sensing unit, including a light intensity sensor, is used to collect ambient light data.

3. The system according to claim 1 or 2, characterized in that, The perception layer device includes at least an AI video acquisition unit, and the edge computing node includes a people flow analysis engine for performing the following operations: Receive a continuous video stream of the target area sent by the AI ​​video acquisition unit; Multiple pedestrians in the continuous video stream are identified using a pre-set target detection model, and a multi-target tracking algorithm is called to assign a unique ID to each pedestrian and record the movement trajectory and speed of each pedestrian to obtain the tracking results; Based on the tracking results, the distribution of the number of people, inbound and outbound traffic, crowd density, average movement speed, and dwell time in the target area is calculated and output in real time.

4. The system according to claim 3, characterized in that, The edge computing node also includes an early warning response engine for: Receive the crowd density output by the crowd flow analysis engine; The population density is compared with a preset density grading threshold, and the comparison result is obtained. The density grading threshold includes at least a first density threshold and a second density threshold. When the comparison result indicates that the population density does not exceed the first density threshold, the target area continues to be monitored by the AI ​​video acquisition unit. When the comparison result indicates that the population density exceeds the first density threshold but does not exceed the second density threshold, a first warning instruction is generated and sent to the application service layer. When the comparison result indicates that the population density exceeds the second density threshold, a second early warning instruction and an emergency response instruction are generated and sent.

5. The system according to claim 1 or 2, characterized in that, The perception layer device includes at least a passenger flow perception unit, and the cloud-based big data platform includes a business intelligence analysis module, which includes: The store value assessment submodule is used to calculate the value index of each store based on historical and real-time foot traffic data and store location attributes through a weighted multi-factor model, and generate a dynamically updated store value map. The merchant operation analysis submodule is used to analyze the merchant's store entry rate and customer dwelling behavior based on the data collected by the customer flow perception unit and the authorized video data in the store, and to generate periodic operation diagnosis reports. The business mix optimization submodule is used to analyze the overall business distribution and customer consumption path of the pedestrian street, identify the complementarity and conflict of business types, and provide suggestions for business type adjustment.

6. The system according to claim 5, characterized in that, The calculation formula for the weighted multi-factor model is as follows: Store Value Index = W1 × Average Daily Customer Traffic + W2 × Customer Dwell Time + W3 × Complementarity Score of Surrounding Businesses + W4 × Accessibility Score + W5 × Visibility Score; Among them, W1, W2, W3, W4, and W5 are the weight coefficients corresponding to the average daily passenger flow, passenger dwell time, complementaryness score of surrounding business formats, accessibility score, and visibility score, respectively, and W1+W2+W3+W4+W5=1.

7. The system according to claim 1, characterized in that, The application service layer includes: The intelligent guidance service module is used to provide tourists with optimal route planning based on real-time location and destination, AR real-scene navigation, barrier-free route planning, personalized shop recommendations, and intelligent parking guidance services. The management cockpit module is used to provide the management terminal with a visually integrated dashboard that includes real-time passenger flow heat maps, early warning information, merchant rankings, facility status, and energy consumption data. The marketing campaign management module provides full-cycle support for marketing campaigns, including pre-campaign traffic forecasting and resource allocation suggestions, real-time monitoring dashboards during the campaign, and multi-dimensional performance evaluation and improvement suggestions after the campaign.

8. The system according to claim 7, characterized in that, The personalized store recommendation service of the intelligent guidance service module is implemented based on user profiles, which are constructed through at least one of the following methods: Preference tags actively selected by users on the tourist terminal; Based on user authorization, the user's age group and gender information are estimated through AI video analysis; The user's search and browsing history.

9. The system according to claim 7, characterized in that, The perception layer device includes at least an AI video acquisition unit and a passenger flow perception unit. The process by which the management cockpit module generates and displays a real-time passenger flow heat map includes: Real-time pedestrian flow data and location data are obtained from the AI ​​video acquisition unit and the pedestrian flow sensing unit deployed at various monitoring points in the pedestrian street at preset time intervals. Based on the location data and the corresponding real-time pedestrian flow, a continuous pedestrian density distribution grid is generated on the pedestrian street plan using a spatial interpolation algorithm. The density values ​​in the crowd density distribution grid are mapped to a preset color spectrum to generate a visual heat map in which the density of crowds is represented by the color depth. Different shades of color correspond to different warning density levels. The visualized heat map is displayed on the visualized integrated dashboard in the management cockpit and is dynamically refreshed according to the preset time interval.

10. The system according to claim 1, characterized in that, The perception layer device includes at least an AI video acquisition unit, and the system further includes a security management engine, which is deployed on the edge computing node or the cloud big data platform and is used for: The AI ​​video acquisition unit analyzes the continuous video stream to monitor the obstruction of fire lanes in real time and automatically generates an alarm when continuous obstruction is detected. Detect abnormal crowd gathering patterns and trigger emergency response procedures; In response to an authorized face search request, face comparison and trajectory tracking are performed in historical video data.

11. The system according to claim 1, characterized in that, The application service layer interacts with users through the following terminals: The large touch screens deployed at the main entrance of the pedestrian street are used to display 3D maps, provide route planning starting point selection, and generate QR codes; A dedicated application on the tourist terminal is used to receive AR navigation, personalized recommendations, special offers, and parking guidance services; The management terminal is used to receive early warning notifications, view cockpit data, and process work orders.

12. A method for comprehensive management of pedestrian streets, applied to the comprehensive management system for pedestrian streets according to any one of claims 1-11, characterized in that, The method includes: Multi-source environmental data of the pedestrian street is continuously collected through sensing layer devices; The collected data is processed in real time at the edge computing node to generate preliminary analysis results, including pedestrian density and trajectory, and to execute local early warnings. The preliminary analysis results and historical data are uploaded to a cloud-based big data platform for in-depth data analysis, model training, and business intelligence decision-making. Through the application service layer, the results of in-depth data analysis and business intelligence decision-making are transformed into management instructions for managers and intelligent guidance services for tourists, thus forming a closed management loop.

13. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer-readable instructions, which, when executed by one or more processors, cause the one or more processors to perform the steps of the pedestrian street integrated management method as described in claim 12.

14. A computer device, characterized in that, include: One or more processors, and memory; The memory stores computer-readable instructions, which, when executed by the one or more processors, perform the steps of the pedestrian street integrated management method as described in claim 12.

Images