Method and system for extracting population activity quantity based on cloud computing and big data utilization

Abstract

Embodiments of the present invention relate to the technical field of population activity monitoring, specifically disclosing a method and system for extracting population activity data based on cloud computing and big data utilization. The embodiments of the present invention collect and upload multi-source population data for a target monitoring area; standardize and fuse the multi-source population data based on cloud computing technology; predict regional population numbers and dynamically visualize them; determine multiple population warning rules, assess the risk of abnormal population numbers in a region, and, when abnormal risks exist, issue warning push notifications and initiate a coordinated response. The present invention achieves precise capture of silent and falsely reported anomalies by constructing a dynamic fusion mechanism that combines an emotion-traffic coupling anomaly heat map with a traffic mutation grid map with conflict indicators.

Description

Technical Field

[0001] The present invention belongs to the technical field of population activity monitoring, and in particular relates to a method and system for extracting the quantity of population activities based on cloud computing and big data utilization. Background Art

[0002] Population activity monitoring is the process of using modern information technology to continuously and systematically monitor and analyze dynamic activities such as the spatial distribution, flow characteristics, density changes and behavioral patterns of the population in a certain area. It comprehensively uses multi-source data such as mobile communication signaling data, satellite remote sensing images, video surveillance, face recognition, geographic information system (GIS), social media data, etc., and through big data analysis, artificial intelligence algorithms and visualization methods, it accurately reflects the population concentration level, migration routes, commuting patterns, population composition and changing trends in a certain area at different time scales.

[0003] Population activity monitoring can not only be applied to urban planning, traffic management, emergency response, public safety and other fields, but also provide data support and scientific basis for major event security, abnormal prevention and control, commercial site selection, etc.

[0004] Existing population activity monitoring technologies lack a unified platform for integrating data, analyzing results, and managing responses, and are usually post-analyzed. They cannot make timely and effective adaptive judgments based on real-time data, and are unable to achieve real-time monitoring and early warning of dynamic changes. Summary of the Invention

[0005] The purpose of the embodiments of the present invention is to provide a method and system for extracting the number of population activities based on cloud computing and big data utilization, aiming to solve the problems raised in the background technology.

[0006] To achieve the above objectives, the embodiments of the present invention provide the following technical solutions:

[0007] A method for extracting population activity data based on cloud computing and big data utilization, the method specifically comprising the following steps:

[0008] Determine a target monitoring area for population monitoring and determine multiple data collection sources, and collect and upload multi-source population data of the target monitoring area according to the multiple data collection sources;

[0009] Based on cloud computing technology, data standardization and fusion processing are performed on the multi-source population data to generate fused population data;

[0010] Identify the fused population data, predict the regional population, and dynamically visualize the regional population;

[0011] Determine multiple population warning rules, make abnormal risk judgments on the population size in the area, and push warnings and respond in a linked manner when there is abnormal risk.

[0012] A population activity quantity extraction system based on cloud computing and big data utilization includes a multi-source collection cloud upload unit, a fusion data processing unit, a population prediction display unit, and an early warning push response unit, wherein:

[0013] A multi-source collection cloud uploading unit is used to determine a target monitoring area for population monitoring and multiple data collection sources, and to collect and upload multi-source population data of the target monitoring area according to the multiple data collection sources;

[0014] A fusion data processing unit, configured to perform data standardization and fusion processing on the multi-source population data based on cloud computing technology to generate fused population data;

[0015] A population prediction and display unit, configured to identify the fused population data, predict the regional population, and dynamically visualize the regional population;

[0016] The early warning push response unit is used to determine multiple population early warning rules, make abnormal risk judgments on the population size in the area, and push early warnings and make linkage responses when there are abnormal risks.

[0017] Compared with the prior art, the present invention has the following beneficial effects:

[0018] The present invention achieves accurate capture of silent anomalies and false-reported anomalies by constructing a dynamic fusion mechanism of the emotion-traffic coupling anomaly heat map and the traffic mutation grid map with conflict indicators; and significantly improves the quality of multi-source population data through a three-level anomaly fuse mechanism and spatial interpolation compensation.

[0019] The present invention realizes the intelligent linkage of the three major charts in the geographic grid dimension through the gridded spatiotemporal linkage mechanism of the line chart mutation points and the stacked chart migration arrows, and intuitively presents the changing characteristics of population activities.

[0020] The present invention constructs a grid-based risk identification framework through dual-source collaborative verification of emotional fuse risk labels and abnormal migration risk labels, combined with video-assisted verification and historical rule matching, to significantly improve the accuracy of abnormal risk judgment.

[0021] The present invention realizes a leapfrog upgrade from static thresholds to dynamic warning rules through dual-source coupling of emotional risk vectors and migration behavior risk tables, combined with dynamic matching of historical rule bases. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are only some embodiments of the present invention.

[0023] Figure 1 A flow chart of a method provided by an embodiment of the present invention is shown.

[0024] Figure 2 The flowchart of multi-source population data collection and uploading in the method provided by the embodiment of the present invention is shown.

[0025] Figure 3 A flow chart of generating fused population data in the method provided by an embodiment of the present invention is shown.

[0026] Figure 4 A flowchart of predicting regional population in the method provided by an embodiment of the present invention is shown.

[0027] Figure 5 A flowchart of the dynamic visualization display of regional population in the method provided by an embodiment of the present invention is shown.

[0028] Figure 6 A flowchart of early warning push and linkage response in the method provided by an embodiment of the present invention is shown.

[0029] Figure 7 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0030] Figure 8 The structure block diagram of the multi-source acquisition cloud upload unit in the system provided by an embodiment of the present invention is shown.

[0031] Figure 9 The figure shows a structural block diagram of a fusion data processing unit in a system provided by an embodiment of the present invention.

[0032] Figure 10 The following is a structural block diagram of an early warning push response unit in a system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0033] In order to make the purpose, technical solutions and advantages of the present invention more clearly understood, the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0034] It is understandable that the population activity monitoring in existing technologies lacks a unified platform to integrate data, analyze results and manage responses, and is usually a post-analysis. It cannot make timely and effective adaptive judgments based on real-time data, and cannot achieve real-time monitoring and early warning of dynamic changes.

[0035] To solve the above problems, the embodiments of the present invention determine the target monitoring area for population monitoring and multiple data collection sources. According to the multiple data collection sources, multi-source population data of the target monitoring area is collected and uploaded; based on cloud computing technology, the multi-source population data is standardized and fused to generate fused population data; the fused population data is identified, the regional population is predicted, and the regional population is dynamically visualized; multiple population warning rules are determined, the regional population is judged for abnormal risks, and when there is an abnormal risk, warning push and linkage response are performed. The system is capable of performing data standardization and fusion processing on multi-source population data, predicting the regional population, and performing dynamic visualization, and when there is an abnormal risk, warning push and linkage response are performed, realizing unified processing of integrated data, analysis results and management response, and can make timely and effective adaptive judgments, realizing real-time monitoring and early warning of dynamic changes.

[0036] Figure 1 A flow chart of a method provided by an embodiment of the present invention is shown.

[0037] Specifically, a method for extracting population activity data based on cloud computing and big data utilization includes the following steps:

[0038] Step S101 : determining a target monitoring area for population monitoring and determining a plurality of data collection sources; collecting and uploading multi-source population data of the target monitoring area according to the plurality of data collection sources.

[0039] In an embodiment of the present invention, by determining the target monitoring area for population monitoring and determining multiple data collection sources including mobile communication sources, terminal application sources, public monitoring sources and social media sources, multi-source population data including mobile communication data (operator's base station positioning data, which can extract the number of users and location changes), terminal application data (anonymous user activity data of map navigation terminal applications), public monitoring data (city public camera monitoring data) and social media data (location information disclosed on social platforms such as Weibo and Douyin) of the target monitoring area are collected through multiple data collection sources, and then the multi-source population data is uploaded to the cloud.

[0040] Specifically, Figure 2 The flowchart of multi-source population data collection and uploading in the method provided by the embodiment of the present invention is shown.

[0041] In a preferred embodiment of the present invention, determining a target monitoring area for population monitoring and determining multiple data collection sources, and collecting and uploading multi-source population data of the target monitoring area according to the multiple data collection sources specifically include the following steps:

[0042] Step S1011, determining the target monitoring area for population monitoring;

[0043] Step S1012, determining multiple data collection sources, including mobile communication sources, terminal application sources, public monitoring sources, and social media sources;

[0044] Step S1013, collecting multi-source population data of the target monitoring area according to the multiple data collection sources;

[0045] Step S1014: Upload the multi-source population data to the cloud.

[0046] Furthermore, the method for extracting the number of population activities based on cloud computing and big data utilization further includes the following steps:

[0047] Step S102: Based on cloud computing technology, the multi-source population data is standardized and integrated to generate integrated population data.

[0048] In an embodiment of the present invention, based on cloud computing technology, data standardization processing is performed on multi-source population data (data from different sources are converted into a unified structure) to obtain multi-source standard data, and then the multi-source standard data is temporally and spatially aligned and calibrated (spatial and temporal alignment is performed based on a unified geographic grid and time step) to obtain multi-source aligned data. Thereafter, anomalies are identified and eliminated from the standard multi-source data (using algorithms such as IQR and Z-score to identify and eliminate unreasonable position changes and noise) to obtain multi-source valid data, and weighted aggregation or Bayesian reasoning is used to fuse the multi-source valid data to generate fused population data.

[0049] Specifically, Figure 3 A flow chart of generating fused population data in the method provided by an embodiment of the present invention is shown.

[0050] In a preferred embodiment of the present invention, the data standardization and fusion processing of the multi-source population data based on cloud computing technology to generate fused population data specifically includes the following steps:

[0051] Step S1021: Based on cloud computing technology, the multi-source population data is subjected to data standardization processing to obtain multi-source standard data;

[0052] Step S1022, performing spatiotemporal alignment and calibration on the multi-source standard data to obtain multi-source aligned data;

[0053] Step S1023, identifying and eliminating anomalies on the multi-source aligned data to obtain multi-source valid data;

[0054] Step S1024: weighted aggregation or Bayesian reasoning is used to fuse the multi-source valid data to generate fused population data.

[0055] Among them, in the preferred embodiment provided by the present invention, the abnormal identification and elimination of the standard multi-source data to obtain the multi-source valid data specifically includes the following steps:

[0056] Step S10231, gridding the multi-source population data in the multi-source aligned data, associating an independent time window label with a data source fingerprint for each grid, and generating a spatiotemporal gridded data cube;

[0057] Step S10232: Obtain the population growth rate for the same period in history. Based on the spatiotemporal grid data cube, compare the population change rate of adjacent time periods, grid by grid, and compare it with the population growth rate for the same period in history to obtain the population growth rate. Mark the traffic mutation grids according to the interval in which the population growth rate falls. Simultaneously record the conflict rate between mobile communication and terminal application data to obtain a traffic mutation grid map with conflict indicators.

[0058] Step S10233: Based on the traffic mutation grid map with conflict indicators, the trajectory points and social media check-in behaviors in the grid are extracted, and anomaly determination is performed based on the time period of the trajectory points in the grid to generate an individual behavior deviation anomaly list. The anomaly determination includes: when the time period of the trajectory point in the grid exceeds the preset period but there is no social media activity, it is determined to be a silent anomaly point; when the social media check-in behavior is intensive but there is no terminal trajectory, it is determined to be a false report anomaly point.

[0059] Step S10234: Based on the list of individual behavioral deviation anomalies, crawl the social media text within the corresponding spatiotemporal grid and match panic keywords using the sentiment dictionary. When the matched panic semantic density exceeds a preset density threshold and overlaps with the traffic mutation grid map with conflict indicators in time and space, it is marked as a sentiment conflict hotspot to construct a sentiment-traffic coupling anomaly heat map.

[0060] Step S10235: Superimpose the individual behavior deviation anomaly list and the emotion-traffic coupling anomaly heat map, and trigger the three-level anomaly fuse mechanism based on the confidence interval of the corresponding data source to obtain the anomaly fuse instruction set;

[0061] Step S10236: Based on the abnormal fuse instruction set, double verification is performed on suspected abnormal points in the abnormal fuse instruction set to generate a clean data pool for the first round of verification. The double verification of suspected abnormal points in the instruction set includes: if the mobile communication data point marked in the instruction set has a corresponding crowd image in a public camera, it is downgraded to a warning item; if the trajectory point has no video evidence and is located in the core area of ​​emotional conflict, real-time elimination is initiated;

[0062] Step S10237: Obtain the mean of the historical grid residual volatility. Based on the clean data pool from the first round of verification, calculate the population residual of each grid after removing outliers, and compare it with the mean of the historical grid residual volatility. Perform spatial interpolation compensation on grids with excessive residuals to form an enhanced dataset for residual compensation.

[0063] Step S10238: Based on the enhanced data set with residual compensation, the population distribution matrix is ​​reconstructed by grid, and data quality labels are added to obtain fused data, which is used as multi-source valid data. The added data quality labels include the anomaly rejection ratio and compensation coverage.

[0064] In this embodiment, the present invention cuts standardized multi-source population data into spatiotemporal gridded data cubes and compares population growth rates grid by grid, and combines mobile communication and terminal application data conflict rates to generate a traffic mutation grid map with conflict indicators; extracts terminal trajectory points within the grid and social media behaviors to determine silent anomalies and false-reported anomalies, and constructs an individual behavior deviation anomaly list; crawls social media texts to match panic keyword density to mark emotional conflict hotspots, and forms an emotion-traffic coupling anomaly heat map; superimposes the anomaly list and heat map to trigger a three-level anomaly fuse mechanism to obtain an anomaly fuse instruction set; verifies suspected anomalies to generate a clean data pool; performs spatial interpolation based on the clean data pool to compensate for residuals, and forms an enhanced data set for residual compensation; and finally reconstructs the population distribution matrix to obtain multi-source valid data.

[0065] Furthermore, the method for extracting the number of population activities based on cloud computing and big data utilization further includes the following steps:

[0066] Step S103: Identify the fused population data, predict the regional population, and dynamically visualize the regional population.

[0067] In an embodiment of the present invention, the fused population data is imported into a preset population estimation model, and the regional population is predicted by the population estimation model. The regional population is then mapped to a preset geographic map. Based on the geographic map, the regional population is rendered to generate heat maps, trajectory flow maps, density cloud maps, etc., and the heat maps, trajectory flow maps, density cloud maps, etc. are visualized. At the same time, combined with time series, the changes in the population of multiple regions are analyzed to generate line graphs, stacked graphs, and radar graphs. Based on the line graphs, stacked graphs, and radar graphs, a visualization of population activity trends is performed.

[0068] Specifically, Figure 4 A flowchart of predicting regional population in the method provided by an embodiment of the present invention is shown.

[0069] In a preferred embodiment of the present invention, identifying the integrated population data, predicting the regional population, and dynamically visualizing the regional population specifically include the following steps:

[0070] Step S1031, importing the fused population data into a preset population estimation model;

[0071] Step S1032, predicting the regional population using the population estimation model;

[0072] Step S1033: Dynamically visualize the population of the region.

[0073] Specifically, Figure 5 A flowchart of the dynamic visualization display of regional population in the method provided by an embodiment of the present invention is shown.

[0074] In a preferred embodiment of the present invention, the dynamic visualization of the population of the region specifically includes the following steps:

[0075] Step S10331, mapping the population of the region to a preset geographical map;

[0076] Step S10332: Render the population of the region based on the geographic map, and generate and visualize a heat map, a trajectory flow map, and a density cloud map;

[0077] Step S10333, performing a change analysis on the population of the plurality of regions, and generating a line chart, a stacked chart, and a radar chart;

[0078] Step S10334: Visually display the trend of population activities based on the line chart, the stacked chart, and the radar chart.

[0079] In a preferred embodiment of the present invention, the analysis of population changes in a plurality of regions to generate a line chart, a stacked chart, and a radar chart specifically includes the following steps:

[0080] Step S103331: Based on valid data from multiple sources, extract the population change curve within a preset time period, synchronously retrieve historical data from the same period to establish an initial baseline, and calibrate the baseline deviation using the data source confidence level of the abnormal fuse instruction set to obtain a calibrated population baseline library;

[0081] Step S103332, based on the enhanced dataset with residual compensation, aggregate population by geographic grid to generate a grid-population aggregation matrix;

[0082] Step S103333, visualizing the calibrated population baseline database and constructing a line graph entity containing mutation points;

[0083] Step S103334, performing cross-grid migration analysis on the grid-population aggregation matrix and constructing a stacked graph entity containing migration arrows;

[0084] Step S103335, constructing an initial radar chart entity based on the population deviation rate of the grid-population aggregation matrix;

[0085] Step S103336: Associate the line chart entity containing the mutation point with a specific geographic grid, so that the migration arrow of the stacked chart entity automatically points to the corresponding grid, and mark the deviation warning area of ​​the corresponding grid in the initial radar chart entity to generate a linked chart group;

[0086] Step S103337: Based on the linked chart group, the chart is split and finally used as a line chart, a stacked chart and a radar chart respectively.

[0087] In this embodiment, the present invention extracts a population change curve through multi-source valid data, and uses an abnormal fuse instruction set to calibrate the historical baseline to obtain a calibrated population baseline library; generates a grid-population aggregation matrix based on an enhanced data set with residual compensation; constructs a line chart entity containing mutation points and a stacked chart entity containing migration arrows; and by associating the line chart mutation points with the geographic grid, the stacked chart migration arrows automatically point to the corresponding grid and mark the deviation warning area on the radar chart, thereby realizing the linkage of the three major charts and solving the problem of chart fragmentation in traditional visualization.

[0088] Furthermore, the method for extracting the number of population activities based on cloud computing and big data utilization further includes the following steps:

[0089] Step S104: determine multiple population warning rules, make an abnormal risk assessment on the population size in the area, and push warnings and perform linkage responses when there is an abnormal risk.

[0090] In an embodiment of the present invention, multiple population warning rules are determined (for example, population density, population growth rate, etc.), and the regional population is analyzed according to the multiple population warning rules to determine whether there is an abnormal risk. If it is determined that there is an abnormal risk, an abnormal warning signal is generated. Based on the abnormal warning signal, multi-channel warning push is carried out (the abnormal warning signal is pushed to the management personnel through SMS, management platform pop-up windows, emails, etc.), and a linkage response is carried out (connected with traffic control and emergency response systems to realize linkage operations such as crowd guidance and area blockade).

[0091] Specifically, Figure 6 A flowchart of early warning push and linkage response in the method provided by an embodiment of the present invention is shown.

[0092] Among them, in the preferred embodiment provided by the present invention, the determination of multiple population warning rules, the abnormal risk judgment of the population size in the area, and the push of warning and linkage response when there is an abnormal risk specifically include the following steps:

[0093] Step S1041, determining multiple population warning rules;

[0094] Step S1042: extracting real-time panic semantic density based on the emotion-traffic coupling anomaly heat map. When the panic semantic density exceeds the standard and the circuit breaker level exceeds the preset level threshold, generating an emotion circuit breaker risk label;

[0095] Step S1043: Analyze the geographic grids pointed by the migration arrows in the stacked graph, and generate an abnormal migration risk label when the migration intensity exceeds a preset intensity threshold;

[0096] Step S1044: retrieve public detection videos to identify changes in the crowd density gradient within the emotional fuse risk label. When the crowd density growth rate matches the migration direction of the abnormal migration risk label, generate a video verification risk event table.

[0097] Step S1045: Backtracking historical events of similar grids based on the enhanced dataset with residual compensation, comparing population change curves, emotional fuse risk labels, and abnormal migration risk labels of historical events of similar grids, and generating a historical confirmation coefficient when the matching degree exceeds a matching threshold;

[0098] Step S1046: When the emotional fuse risk tag and the abnormal migration risk tag point to the same geographic grid, and the video confirms that they exist in the risk event table, and the historical confirmation coefficient exceeds the preset coefficient threshold, it is determined to be a regional linkage abnormal risk, and a determination result is obtained;

[0099] Step S1047, based on the determination result, determine whether there is an abnormal risk; when there is an abnormal risk, generate an abnormal warning signal; according to the abnormal warning signal, push warnings through multiple channels; according to the abnormal warning signal, perform a linkage response.

[0100] In this embodiment, the present invention generates an emotion fuse risk label through an emotion-traffic coupling anomaly heat map; generates an abnormal migration risk label by analyzing the migration arrows of the stacked map; retrieves the public video verification density growth rate and matches it with the migration direction to generate a video verification event; generates a historical verification coefficient based on the residual compensation data set to trace back historical events; when the emotion and migration risk point to the same grid and there is video verification and historical verification exceeding the threshold, the regional linkage abnormal risk is determined to construct a multi-source collaborative verification framework.

[0101] In a preferred embodiment of the present invention, determining a plurality of population warning rules specifically includes the following steps:

[0102] Step S10411: extracting emotional conflict hotspots based on the emotion-traffic coupling anomaly heat map and generating an emotional risk vector;

[0103] Step S10412: Analyze the direction and strength of the migration arrows in the stacked graph entity to generate a migration behavior risk correspondence table;

[0104] Step S10413: When the emotional risk vector and the migration behavior risk correspondence table point to the same geographical grid, it is marked as a dual-source coupling risk event;

[0105] Step S10414: Backtracking historical grid events of the same type based on the enhanced data set with residual compensation, extracting handling rules, and generating a historical rule base;

[0106] Step S10415, matching the threshold of the corresponding grid type in the historical rule library according to the dual-source coupling risk event to generate a rule trigger result;

[0107] Step S10416: When the rule triggering result exceeds the preset rule threshold, a cross-grid linkage warning rule is generated, and the cross-grid linkage warning rule is used as a population warning rule.

[0108] In this embodiment, the present invention generates an emotion risk vector through an emotion-traffic coupling anomaly heat map; generates a migration behavior risk table through stacked graph entities; marks a dual-source coupling risk event when both point to the same grid; generates a historical rule base based on a residual compensation data set by backtracking historical events; generates a rule trigger result for the risk event matching the rule base threshold; and generates a cross-grid linkage warning rule when the trigger result exceeds the threshold, thereby solving the defect of insufficient adaptability of traditional warning rules.

[0109] Further, Figure 7 The application architecture diagram of the system provided by the embodiment of the present invention is shown.

[0110] Among them, in another preferred embodiment provided by the present invention, the population activity quantity extraction system based on cloud computing and big data utilization includes:

[0111] The multi-source collection cloud uploading unit 101 is used to determine a target monitoring area for population monitoring and multiple data collection sources, and collect and upload multi-source population data of the target monitoring area according to the multiple data collection sources.

[0112] In an embodiment of the present invention, the multi-source collection cloud upload unit 101 determines the target monitoring area for population monitoring and determines multiple data collection sources including mobile communication sources, terminal application sources, public monitoring sources and social media sources. Then, through the multiple data collection sources, it collects multi-source population data of the target monitoring area including mobile communication data (operator's base station positioning data, which can extract the number of users and location changes), terminal application data (anonymous user activity data of map navigation terminal applications), public monitoring data (city public camera monitoring data) and social media data (location information disclosed on social platforms such as Weibo and Douyin), and then uploads the multi-source population data to the cloud.

[0113] Specifically, Figure 8 It shows a structural block diagram of the multi-source acquisition cloud upload unit 101 in the system provided by an embodiment of the present invention.

[0114] In the preferred embodiment of the present invention, the multi-source acquisition cloud upload unit 101 specifically includes:

[0115] An area determination module 1011 is used to determine a target monitoring area for population monitoring;

[0116] a source determination module 1012 for determining multiple data collection sources, including mobile communication sources, terminal application sources, public monitoring sources, and social media sources;

[0117] A data collection module 1013 is configured to collect multi-source population data of the target monitoring area according to the plurality of data collection sources;

[0118] The data cloud uploading module 1014 is used to upload the multi-source population data to the cloud.

[0119] Furthermore, the population activity quantity extraction system based on cloud computing and big data utilization also includes:

[0120] The fusion data processing unit 102 is used to perform data standardization and fusion processing on the multi-source population data based on cloud computing technology to generate fused population data.

[0121] In an embodiment of the present invention, the fusion data processing unit 102 performs data standardization processing on multi-source population data based on cloud computing technology (converting data from different sources into a unified structure) to obtain multi-source standard data, and then performs spatiotemporal alignment and calibration on the multi-source standard data (based on a unified geographic grid and time step, spatial and temporal alignment) to obtain multi-source aligned data. Thereafter, anomalies are identified and eliminated on the standard multi-source data (using algorithms such as IQR and Z-score to identify and eliminate unreasonable position changes and noise) to obtain multi-source valid data, and weighted aggregation or Bayesian reasoning is used to fuse the multi-source valid data to generate fused population data.

[0122] Specifically, Figure 9 FIG. 1 shows a structural block diagram of the fusion data processing unit 102 in the system provided by an embodiment of the present invention.

[0123] In a preferred embodiment of the present invention, the fusion data processing unit 102 specifically includes:

[0124] A standardization processing module 1021 is used to perform data standardization processing on the multi-source population data based on cloud computing technology to obtain multi-source standard data;

[0125] A spatiotemporal alignment and calibration module 1022 is configured to perform spatiotemporal alignment and calibration on the multi-source standard data to obtain multi-source aligned data;

[0126] Anomaly identification and elimination module 1023, used to identify and eliminate anomalies in the multi-source aligned data to obtain multi-source valid data;

[0127] The fusion processing module 1024 is used to perform fusion processing on the multi-source valid data by using weighted aggregation or Bayesian reasoning to generate fused population data.

[0128] Furthermore, the population activity quantity extraction system based on cloud computing and big data utilization also includes:

[0129] The population prediction and display unit 103 is used to identify the fused population data, predict the regional population, and dynamically visualize the regional population.

[0130] In an embodiment of the present invention, the population prediction and display unit 103 imports the integrated population data into a preset population estimation model, predicts the regional population through the population estimation model, and then maps the regional population to a preset geographic map. Based on the geographic map, the regional population is rendered to generate a heat map, a trajectory flow map, a density cloud map, etc., and the heat map, trajectory flow map, density cloud map, etc. are visualized. At the same time, combined with the time series, the population changes in multiple regions are analyzed to generate line graphs, stacked graphs and radar graphs. Based on the line graphs, stacked graphs and radar graphs, a visualization display of population activity trends is performed.

[0131] The early warning push response unit 104 is used to determine multiple population early warning rules, make abnormal risk judgments on the population size in the area, and push early warnings and perform linkage responses when there is an abnormal risk.

[0132] In an embodiment of the present invention, the early warning push response unit 104 determines multiple population early warning rules (for example, population density, population growth rate, etc.), analyzes the regional population according to the multiple population early warning rules, determines whether there is an abnormal risk, and generates an abnormal early warning signal when it is determined that there is an abnormal risk. According to the abnormal early warning signal, multi-channel early warning push is performed (the abnormal early warning signal is pushed to the management personnel through SMS, management platform pop-up windows, emails, etc.), and a linkage response is performed (connected with traffic control and emergency response systems to realize linkage operations such as crowd guidance and area blockade).

[0133] Specifically, Figure 10 It shows a structural block diagram of the early warning push response unit 104 in the system provided by an embodiment of the present invention.

[0134] In a preferred embodiment of the present invention, the warning push response unit 104 specifically includes:

[0135] An early warning rule determination module 1041 is used to determine multiple population early warning rules;

[0136] An abnormal risk judgment module 1042 is used to analyze the population of the area according to the plurality of population warning rules to determine whether there is an abnormal risk;

[0137] The signal generating module 1043 is used to generate an abnormal warning signal when there is an abnormal risk;

[0138] The warning push module 1044 is used to push warnings through multiple channels according to the abnormal warning signal;

[0139] The linkage response module 1045 is used to perform a linkage response according to the abnormal warning signal.

[0140] It should be understood that, although the various steps in the flow chart of each embodiment of the present invention are shown in sequence according to the indication of the arrows, these steps are not necessarily performed in sequence according to the order indicated by the arrows. Unless otherwise specified herein, the execution of these steps is not strictly limited in order, and these steps can be performed in other orders. Moreover, at least a portion of the steps in each embodiment may include a plurality of sub-steps or a plurality of stages, and these sub-steps or stages are not necessarily performed at the same time, but can be performed at different times, and the execution order of these sub-steps or stages is not necessarily performed in sequence, but can be performed in turn or alternately with at least a portion of other steps or sub-steps or stages of other steps.

[0141] Those skilled in the art will appreciate that all or part of the processes in the above-described method embodiments can be implemented by instructing the relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When executed, the program can include the processes of the above-described method embodiments. Any reference to memory, storage, database, or other media used in the various embodiments provided herein may 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. By way of illustration and not limitation, RAM is available in various 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 RAMbus dynamic RAM (DRDRAM), and RAMbus dynamic RAM (RDRAM).

[0142] The technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, not all possible combinations of the technical features in the above-mentioned 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.

[0143] The above-described embodiments merely illustrate several implementations of the present invention, and while their descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that a person skilled in the art would be able to make numerous variations and improvements without departing from the spirit of the present invention, all of which fall within the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be determined by the appended claims.

[0144] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for extracting population activity data based on cloud computing and big data, characterized in that: The method specifically comprises the following steps: Determine a target monitoring area for population monitoring and determine multiple data collection sources, and collect and upload multi-source population data of the target monitoring area according to the multiple data collection sources; Based on cloud computing technology, data standardization and fusion processing are performed on the multi-source population data to generate fused population data; Identify the fused population data, predict regional population numbers, and dynamically visualize the regional population numbers; Determine multiple population warning rules, judge the abnormal risk of the population in the area, and push warnings and respond in a coordinated manner when there is an abnormal risk; The method of performing data standardization and fusion processing on the multi-source population data based on cloud computing technology to generate fused population data specifically includes the following steps: Based on cloud computing technology, data standardization processing is performed on the multi-source population data to obtain multi-source standard data; Performing spatiotemporal alignment and calibration on the multi-source standard data to obtain multi-source aligned data; Identifying and eliminating anomalies on the multi-source aligned data to obtain multi-source valid data; Using weighted aggregation or Bayesian reasoning to fuse the multi-source valid data to generate fused population data; The method of identifying and eliminating anomalies from the standard multi-source data to obtain multi-source valid data specifically includes the following steps: The multi-source population data in the multi-source aligned data are gridded, and each grid is associated with an independent time window label and data source fingerprint to generate a spatiotemporal gridded data cube; Obtain the population growth rate during the same period in history. Based on the spatiotemporal grid data cube, compare the population change rate of adjacent time periods on a grid-by-grid basis and compare it with the population growth rate during the same period in history to obtain the population growth rate. Mark the traffic mutation grids according to the interval in which the population growth rate falls. Simultaneously record the conflict rate between mobile communication and terminal application data to obtain a traffic mutation grid map with conflict indicators. Based on a traffic mutation grid map with conflict indicators, we extract the trajectory points and social media check-in behaviors within the grid, and determine anomalies based on the time periods in which the trajectory points stay within the grid, generating a list of individual behavioral deviation anomalies. Based on a list of individual behavioral deviation anomalies, we crawl social media text within the corresponding spatiotemporal grid and match panic keywords using a sentiment dictionary. When the matched panic semantic density exceeds a preset density threshold and overlaps with the traffic mutation grid map with conflict indicators in time and space, it is marked as an emotional conflict hotspot to construct a heat map of emotion-traffic coupling anomalies. The individual behavior deviation anomaly list and the emotion-traffic coupling anomaly heat map are superimposed, and based on the confidence interval of the data source, the three-level anomaly circuit breaker mechanism is triggered to obtain the anomaly circuit breaker instruction set; Based on the abnormal circuit breaker instruction set, double verification is performed on the suspected abnormal points in the abnormal circuit breaker instruction set to generate a clean data pool for the first round of verification; Obtain the mean of the historical grid residual volatility. Based on the clean data pool verified in the first round, calculate the population residual of each grid after removing outliers and compare it with the mean of the historical grid residual volatility. Perform spatial interpolation compensation on grids with excessive residuals to form an enhanced dataset for residual compensation. Based on the enhanced dataset with residual compensation, the population distribution matrix is ​​reconstructed by grid and data quality labels are added to obtain fused data, which is used as multi-source valid data.

2. The method for extracting population activity quantity based on cloud computing and big data utilization according to claim 1 is characterized in that: Determining a target monitoring area for population monitoring, determining multiple data collection sources, and collecting and uploading multi-source population data of the target monitoring area according to the multiple data collection sources specifically includes the following steps: Identify target monitoring areas for population monitoring; Identify multiple data collection sources, including mobile communication sources, terminal application sources, public monitoring sources, and social media sources; Collect multi-source population data of the target monitoring area according to the multiple data collection sources; The multi-source population data is uploaded to the cloud.

3. The method for extracting population activity quantity based on cloud computing and big data utilization according to claim 1 is characterized in that: The identifying of the fused population data, predicting the regional population, and dynamically visualizing the regional population specifically include the following steps: Importing the fused population data into a preset population estimation model; Predicting regional population using the population estimation model; A dynamic visual display of the population of the area is provided.

4. The method for extracting population activity quantity based on cloud computing and big data utilization according to claim 3 is characterized in that: The dynamic visualization of the population of the region specifically includes the following steps: Mapping the population of the region onto a preset geographical map; Based on the geographic map, the population of the region is rendered to generate and visualize a heat map, a trajectory flow map, and a density cloud map; Analyze the changes in population numbers in multiple areas and generate line charts, stacked charts and radar charts; Based on the line chart, the stacked chart and the radar chart, a visual display of population activity trends is performed.

5. The method for extracting population activity quantity based on cloud computing and big data utilization according to claim 4 is characterized in that: The analysis of the change in population of the plurality of regions to generate a line chart, a stacked chart and a radar chart specifically includes the following steps: Based on valid data from multiple sources, the population change curve within a preset time period is extracted, and historical data from the same period is simultaneously retrieved to establish an initial baseline. The baseline deviation is calibrated using the data source confidence level of the abnormal circuit breaker instruction set to obtain a calibrated population baseline database. Based on the enhanced dataset with residual compensation, the population is aggregated by geographic grid to generate a grid-population aggregation matrix; Visualize the calibrated population baseline library and construct a line graph entity containing mutation points; Conduct cross-grid migration analysis on the grid-population aggregation matrix and construct a stacked graph entity containing migration arrows; Based on the population deviation rate of the grid-population aggregation matrix, the initial radar map entity is constructed; Associate the line chart entity containing the mutation point to a specific geographic grid, so that the migration arrow of the stacked chart entity automatically points to the corresponding grid, and mark the deviation warning area of ​​the corresponding grid in the initial radar chart entity to generate a linked chart group; Based on the linked chart group, the chart is split and finally used as a line chart, stacked chart and radar chart respectively.

6. The method for extracting population activity data based on cloud computing and big data utilization according to claim 5, characterized in that: The method of determining a plurality of population warning rules, judging the abnormal risk of the population in the region, and pushing warnings and performing linkage responses when there is an abnormal risk specifically includes the following steps: Determine multiple population warning rules; Extract real-time panic semantic density based on the emotion-traffic coupling anomaly heat map. When the panic semantic density exceeds the standard and the circuit breaker level exceeds the preset threshold, an emotional circuit breaker risk label is generated. Parse the stacked graph migration arrows pointing to the geographic grid and generate an abnormal migration risk label when the migration intensity exceeds the preset intensity threshold; Retrieve public detection videos to identify changes in crowd density gradients within the emotional fuse risk label. When the crowd density growth rate matches the migration direction of the abnormal migration risk label, generate a video-verified risk event table. The enhanced dataset based on residual compensation looks back at historical events of similar grids, compares the population change curves, emotional fuse risk labels, and abnormal migration risk labels of historical events of similar grids, and generates a historical confirmation coefficient when the matching degree exceeds the matching threshold; When the emotional fuse risk label and the abnormal migration risk label point to the same geographical grid, and the video confirms the existence of the risk event table, and the historical confirmation coefficient exceeds the preset coefficient threshold, it is determined to be a regional linkage abnormal risk and the judgment result is obtained; Based on the judgment result, determine whether there is an abnormal risk; when there is an abnormal risk, generate an abnormal warning signal; according to the abnormal warning signal, push warnings through multiple channels; according to the abnormal warning signal, perform a linkage response.

7. The method for extracting population activity data based on cloud computing and big data utilization according to claim 6, characterized in that: Determining multiple population warning rules specifically includes the following steps: Extract emotional conflict hotspots based on the emotion-traffic coupling anomaly heat map and generate emotional risk vectors; Analyze the direction and strength of the migration arrows in the stacked graph entities and generate a migration behavior risk correspondence table; When the emotional risk vector and the migration behavior risk correspondence table point to the same geographical grid, it is marked as a dual-source coupling risk event; Based on the enhanced data set of residual compensation, historical grid events of the same type are traced back, treatment rules are extracted, and a historical rule base is generated; According to the dual-source coupling risk event, the threshold of the corresponding grid type in the historical rule library is matched to generate the rule trigger result; When the rule triggering result exceeds the preset rule threshold, a cross-grid linkage warning rule is generated and used as a population warning rule.

8. A population activity quantity extraction system based on cloud computing and big data utilization, characterized in that: The system applies the method for extracting population activity data based on cloud computing and big data utilization as described in any one of claims 1 to 7 above, and includes a multi-source collection cloud upload unit, a fusion data processing unit, a population prediction display unit, and an early warning push response unit, wherein: A multi-source collection cloud uploading unit is used to determine a target monitoring area for population monitoring and multiple data collection sources, and to collect and upload multi-source population data of the target monitoring area according to the multiple data collection sources; A fusion data processing unit, configured to perform data standardization and fusion processing on the multi-source population data based on cloud computing technology to generate fused population data; A population prediction and display unit, configured to identify the fused population data, predict the regional population, and dynamically visualize the regional population; The early warning push response unit is used to determine multiple population early warning rules, make abnormal risk judgments on the population size in the area, and push early warnings and make linkage responses when there are abnormal risks.

Images