A Location-Based Outdoor Advertising Placement Intelligent Dispatch and Monitoring Platform

By constructing an outdoor advertising location model and calculating adaptation scores from multi-source data, dynamic geofencing is generated, solving the problems of unreasonable order dispatch and large positioning errors in outdoor advertising monitoring, and realizing intelligent order dispatch and efficient on-site verification.

CN123117680APending Publication Date: 2026-11-17JIAYI INNOVATION (BEIJING) CULTURE TECHNOLOGY CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202611374493.4
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-09-07
Publication Date
2026-11-17

AI Technical Summary

Technical Problem

Existing outdoor advertising monitoring operations suffer from problems such as poor dispatching rationality, long commuting distances for personnel, low operational efficiency, large positioning errors, and low reliability of location verification.

Method used

By constructing a spatiotemporal location model of advertising locations, combining multi-source status data to calculate the monitoring terminal adaptation score, generating dynamic geofences, and combining IMU motion trajectories for on-site verification, a closed-loop automatic workflow of work orders can be achieved.

Benefits of technology

Intelligent dispatching has been implemented, which has improved the rationality of dispatching and operational efficiency, reduced location misjudgment, and enhanced the reliability of on-site location verification.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN123117680A_ABST
    Figure CN123117680A_ABST
Patent Text Reader

Abstract

This invention discloses an intelligent dispatch and monitoring platform for outdoor advertising based on LBS positioning, belonging to the field of outdoor advertising supervision technology. It includes an advertising order access module, which constructs a spatiotemporal location model of advertising sites and establishes topological relationships among densely populated sites; an LBS positioning acquisition module, which collects multi-source status data from monitoring terminals; an aggregation and grouping module, which aggregates and groups spatially adjacent sites into work orders and resolves conflicts for duplicate work orders; an intelligent dispatch and scheduling module, which calculates the adaptation score of each monitoring terminal and issues monitoring work orders based on the adaptation score; and an on-site verification module, which verifies on-site operations and, based on the verification results, performs material interception, evidence storage, work order closure, and generates rectification and review work orders. This solution addresses the problems of unreasonable manual dispatching, inaccurate location verification, susceptibility to fraud in verification, and low automation in work order processing, effectively improving the rationality of outdoor advertising monitoring and dispatching and overall operational efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] This invention relates to the field of outdoor advertising placement monitoring technology, and more specifically, to an intelligent dispatch monitoring platform for outdoor advertising placement based on LBS positioning. Background Technology

[0002] With the rapid development of the outdoor advertising industry, a large number of outdoor advertising locations have been set up in business districts, streets, and buildings. Advertisers need to regularly verify the on-site advertisements to confirm whether the advertisements are published normally according to the order requirements, whether the materials are compliant, and whether the placement time is appropriate. At present, most traditional outdoor advertising monitoring operations adopt a manual offline dispatch mode: the back-end dispatch personnel rely on experience to manually allocate monitoring work orders and manually assign inspection personnel to each advertising location to complete on-site verification.

[0003] Shortcomings of existing technology:

[0004] The work order allocation does not take into account multiple factors such as the geographical location of the inspectors, real-time load, historical work quality, business skills, and on-site road conditions. The work order allocation is not reasonable and is prone to problems such as concentrated task overload, long commuting distance of personnel, and low work efficiency.

[0005] Advertising locations are often obstructed by tall buildings and urban canyons, and GNSS positioning has inherent drift errors. This can easily lead to situations where a real person is mistakenly intercepted, or where location fraud passes verification, resulting in low reliability of location verification.

[0006] To address the above problems, this invention proposes a solution. Summary of the Invention

[0007] To overcome the aforementioned deficiencies of the prior art, embodiments of the present invention provide an intelligent dispatch and monitoring platform for outdoor advertising based on LBS positioning. This platform aggregates and groups advertising work orders, calculates monitoring terminal adaptation scores based on multi-source status data to achieve intelligent hierarchical dispatch, and then adaptively generates dynamic geofences by combining historical GPS error compensation parameters and real-time satellite signal-to-noise ratio. It also uses IMU motion trajectories to determine the authenticity of on-site presence and conducts multimodal material verification and evidence storage, achieving an automatic closed-loop workflow for work orders to solve the problems mentioned in the background art.

[0008] To achieve the above objectives, the present invention provides the following technical solution:

[0009] A location-based (LBS)-based intelligent order dispatch and monitoring platform for outdoor advertising includes the following steps:

[0010] The advertising order access module acquires outdoor advertising placement orders, extracts advertising location information, constructs a spatiotemporal location model of advertising locations, and establishes topological relationships between dense locations.

[0011] The LBS positioning and acquisition module collects multi-source status data from the monitoring terminal, including raw sensing data, monitoring personnel profile parameters, and traffic environment data around the location.

[0012] The aggregation and grouping module aggregates and groups work orders for spatially adjacent points based on the topological relationships of dense points, and resolves conflicts for duplicate work orders.

[0013] The intelligent dispatching module calculates the adaptation score of each monitoring terminal based on multi-source status data for the locations within the current aggregated work order group, and issues monitoring work orders based on the adaptation score.

[0014] The on-site verification module receives advertising verification multimedia materials submitted by the monitoring terminal, and performs on-site verification by combining the terminal's reported location, satellite signal-to-noise ratio, and IMU motion trajectory. Based on the verification results, it performs material interception, evidence storage, work order closure, and generates rectification and review work orders.

[0015] In a preferred embodiment, the process of constructing the spatiotemporal location model of the advertising location is as follows:

[0016] Receive outdoor advertising placement order messages and verify the legality of the order messages;

[0017] After successful verification, extract the advertising location information, including the unique identifier of the location, the reference latitude and longitude of the location, the advertising type, the start time of the advertising campaign, the end time of the advertising campaign, and the reference material sample set for the advertising campaign.

[0018] Read the historical GNSS positioning statistics log of this point, calculate the positioning error of multiple historical positioning results of this point relative to the reference position, calculate the mean and standard deviation of historical positioning error, and combine them to generate GPS error compensation parameters for this point.

[0019] Based on the GPS error compensation parameters of this location, combined with the unique identifier of the location, the reference latitude and longitude of the location, the start time of the advertising campaign, the end time of the advertising campaign, and the reference material sample set of the advertising campaign, a spatiotemporal location model of this location is assembled.

[0020] In a preferred embodiment, the process of establishing the topological associations of dense points is as follows:

[0021] Traverse the entire set of advertising locations already in the database and use the Haversine formula to calculate the spherical geodetic distance between location i and location j.

[0022] Construct a dense topological graph of ad placements, where each vertex in the vertex set corresponds to an ad placement.

[0023] Set a spatial proximity threshold. If the spherical distance between point i and point j is less than or equal to the spatial proximity threshold, then generate an undirected connection edge between vertex i and vertex j in the dense point topology graph, marking point i and point j as spatially dense neighbors. Otherwise, do not establish a connection edge between them, thus obtaining the edge set.

[0024] In a preferred embodiment, the raw sensing data includes the terminal's GNSS positioning, satellite signal-to-noise ratio, and effective motion trajectory.

[0025] In a preferred embodiment, the monitoring personnel profile parameters include historical verification accuracy, the number of work orders currently accepted, and the original skill matching coefficient.

[0026] The historical verification accuracy rate is the percentage of work orders with correct manual review results among all the personnel's historical verification work orders.

[0027] The number of work orders currently accepted is the number of monitoring work orders that the person has not yet completed and are currently in transit.

[0028] The process of obtaining the original skill matching coefficient is as follows:

[0029] Extract the advertising type tag of the current work order to be monitored; read the job tag set of the monitored personnel and determine whether the set contains the advertising type tag; if the tag is not contained, set the original skill matching score to 0, which means that the personnel do not have the qualifications for the job.

[0030] If the tag is included, read the preset tag weight corresponding to that tag;

[0031] Obtain the historical accuracy rate of similar work orders for this person;

[0032] The original skill matching score is calculated based on the corresponding preset label weights and the historical accuracy rate of the personnel's similar types of work orders.

[0033] Perform maximum-min normalization on the original skill matching score and output the normalized skill matching coefficient.

[0034] In a preferred embodiment, the process of aggregating and grouping work orders for spatially adjacent points based on the topological relationships of dense points is as follows:

[0035] Receive the topological relationship map of dense points and the spatiotemporal point model of each point, and read the point ID and deployment time window of each point;

[0036] Receive all multi-source status data packets preprocessed by the online monitoring terminals;

[0037] All pending monitoring work orders are retrieved from the business database, and a set of pending work orders is constructed. Each pending work order is bound to a unique location identifier.

[0038] Traverse the set of pending work orders and find the set of points that are spatially close to each other based on the topological relationship graph;

[0039] For spatially adjacent points, read the deployment time window in their respective spatiotemporal point models and determine whether the time windows overlap.

[0040] Work orders that simultaneously meet the criteria of spatial proximity and overlapping delivery time windows will be grouped into the same aggregated work order group.

[0041] In a preferred embodiment, the conflict resolution process for duplicate work orders is as follows:

[0042] If there are duplicate work orders pending at the same location, read the business priority of each duplicate work order;

[0043] Only the work order with the highest business priority will be retained as a valid work order;

[0044] Other duplicate work orders are marked as redundant work orders, their status is set to archived and voided, and they will not participate in subsequent work order dispatch.

[0045] Update the aggregated work order group and the set of work orders to be processed, remove redundant work orders that have been invalidated, and obtain the set of valid work orders after conflict resolution.

[0046] In a preferred embodiment, the process of calculating the adaptation score of each monitoring terminal based on multi-source status data is as follows:

[0047] Obtain the reference latitude and longitude of all points within the aggregated work order group, and calculate the average latitude and longitude within the aggregated work order group based on the reference latitude and longitude of all points, which serves as the spatial reference center point of the aggregated group.

[0048] Traverse all online monitoring terminals and use the Haversine spherical distance formula to calculate the spherical distance from the current position of the monitoring terminal to the center point of the aggregation group;

[0049] Based on the spherical distance from the current location of the monitoring terminal to the center point of the aggregation group, the distance factor is obtained using reverse maximum-minimum normalization:

[0050] Obtain the number of monitoring work orders that are currently in transit and have not yet been completed by the monitoring personnel, and perform reverse max-min normalization on the number of monitoring work orders in transit to obtain the normalized load factor.

[0051] The normalized skill matching coefficient and congestion factor are obtained from multi-source state data, and the historical verification accuracy is calculated. The matching score is obtained by weighting the normalized load factor and distance factor.

[0052] In a preferred embodiment, the process of issuing a monitoring work order based on the adaptation score is as follows:

[0053] If the spherical distance from the current location of the monitoring terminal to the center point of the aggregation group is less than or equal to the configurable distance filtering threshold, the online monitoring terminal will be included in the first-level candidate pool.

[0054] If the spherical distance from the current location of the monitoring terminal to the center point of the aggregation group is greater than the configurable distance filtering threshold, the monitoring terminal will be classified into the second-level candidate pool.

[0055] If the set of valid monitoring personnel in the first-level candidate pool is not empty, then all monitoring personnel in the first-level candidate pool are sorted in descending order according to their comprehensive suitability scores, and a monitoring work order is issued to the monitoring personnel with the highest suitability score; if the monitoring personnel does not accept the issued work order, then work orders are issued in order according to the comprehensive suitability scores.

[0056] If none of the valid monitoring personnel in the first-level candidate pool have accepted the issued work order or the set of valid monitoring personnel in the first-level candidate pool is empty, then the second-level candidate pool will be activated.

[0057] All monitoring personnel in the second-level candidate pool are sorted in descending order according to their comprehensive fit score. A monitoring work order is issued to the monitoring personnel with the highest fit score in the second-level candidate pool. If the monitoring personnel does not accept the issued work order, work orders are issued in order according to the comprehensive fit score.

[0058] In a preferred embodiment, the on-site verification module is implemented as follows:

[0059] The system receives advertising verification multimedia materials reported by the monitoring terminal, and binds the multimedia materials with the location, trajectory, and signal-to-noise ratio data at the same time to obtain the associated verification dataset.

[0060] Read the GPS error compensation parameters of the advertising location and calculate the basic fence radius of the location according to the 3σ principle;

[0061] Obtain the average satellite signal-to-noise ratio (SNR), and normalize the average satellite SNR according to the preset upper and lower limits of SNR to obtain the normalized SNR value.

[0062] The normalized signal-to-noise ratio is used to generate a scaling factor to correct the basic fence radius of the point location, thus obtaining the dynamic geofence radius;

[0063] Using the base geographical location of the advertising location as the center, draw a circle with the radius of the dynamic fence to form a dynamic geofence;

[0064] Based on the IMU motion trajectory, a comprehensive judgment is made as to whether the monitoring terminal has actually arrived at the dynamic geofence area corresponding to the location; if it is determined that there is location spoofing or location drift, the multimedia material is intercepted for verification, and the work order remains in the pending verification state; after the location verification is passed, the material verification process begins.

[0065] Based on the comprehensive location verification results, the matching degree of advertising materials, the consistency of the on-site background, and the matching of shooting time, the verification confidence level is output.

[0066] The verification confidence level is compared with the preset verification. When the verification confidence level is greater than or equal to the verification pass threshold, the verification is determined to be passed. The location data, motion trajectory, multimedia materials, and timestamp are packaged into a data package for evidence storage. The system marks the work order that needs to be reviewed and transfers it to the manual review queue.

[0067] When the verification confidence level is less than the verification pass threshold, the verification is deemed unsuccessful, and a rectification and review work order is generated and sent to the monitoring terminal. After the monitoring terminal resubmits the verification materials, the verification steps are executed again. After the second verification is completed, the work order is closed based on the verification results.

[0068] The technical effects and advantages of the LBS-based intelligent order dispatch monitoring platform for outdoor advertising placement in this invention are as follows:

[0069] 1. This invention constructs aggregated work order groups and calculates the comprehensive adaptation score of monitoring terminals by integrating multi-source status data. It integrates multiple dimensions of indicators such as geographical location, real-time traffic conditions, personnel load, historical verification quality, and business skills to achieve automated intelligent work order dispatch, abandoning the traditional manual experience-based work order dispatch mode. It prioritizes the monitoring terminal with the best comprehensive conditions to undertake tasks, and adds load prediction and unpacking allocation mechanisms to avoid personnel overload, shorten inspection commuting distance, and effectively improve the rationality of outdoor advertising monitoring work order dispatch and overall operational efficiency.

[0070] 2. This invention adaptively generates a dynamic geofence by combining historical GPS error compensation parameters of the location with the average satellite signal-to-noise ratio reported by the terminal. It dynamically adjusts the fence radius according to the inherent positioning deviation of different locations and the real-time satellite signal quality, and then uses IMU motion trajectory to jointly determine whether the terminal has actually arrived at the location area. It overcomes the defect of fixed radius geofences that are prone to misjudgment in urban canyons and building obstruction scenarios. It not only reduces false interception of actual arrivals, but also can identify positioning drift and positioning spoofing behavior, which greatly improves the reliability of on-site location verification. Attached Figure Description

[0071] Figure 1 This is a schematic diagram of the structure of an intelligent dispatch monitoring platform for outdoor advertising based on LBS positioning according to the present invention. Detailed Implementation

[0072] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0073] Example 1, Figure 1 This invention presents an intelligent order dispatch monitoring platform for outdoor advertising based on LBS positioning.

[0074] The advertising order access module acquires outdoor advertising placement orders, extracts advertising location information, constructs a spatiotemporal location model of advertising locations, and establishes topological relationships between dense locations.

[0075] Receive outdoor advertising placement order messages from the business platform and verify the legality of the order messages; after successful verification, extract the advertising location information, including the unique identifier of the location. Point reference latitude and longitude Ad type, ad launch start time End time of advertising campaign Advertising benchmark material sample set ;

[0076] Read the historical GNSS positioning statistics log for this location, calculate the positioning error of multiple historical positioning results relative to the reference position, and obtain the average historical positioning error. Error standard deviation Combined generation of GPS error compensation parameters for this location ;

[0077] Combined with unique location identifier Point reference latitude and longitude The start time of advertising campaign End time of advertising campaign Advertising benchmark material sample set Assemble to form a spatiotemporal point model of that location. ;

[0078] in It is a unique identifier for each advertising location; It is the reference latitude and longitude of the location; These are GPS error compensation parameters; It is the advertising placement time window; It is a sample set of advertising benchmark materials; the spatiotemporal location model of each location is persistently stored in the database.

[0079] Iterate through the entire set of existing ad placements and use the Haversine formula to calculate the placements. with point The spherical distance between The formula is:

[0080] ;

[0081] In the formula, The equivalent radius of the Earth; For the first The latitude (radians) of each advertising location. For the first The latitude (radians) of each advertising location. For the first The longitude (in radians) of each advertising location. For the first The longitude (in radians) of each advertising location.

[0082] The algorithm employs the Haversine spherical distance formula to calculate the distance between two points on the Earth's great circle surface based on two sets of GPS latitude and longitude coordinates. Before calculation, latitude and longitude in angular form need to be converted to radian form for calculation. This algorithm has good numerical stability for calculating short-distance points in cities and is suitable for scenarios such as spatial clustering of advertising locations and geographic distance verification. By approximating the Earth as a standard sphere, it can meet the accuracy requirements of outdoor advertising monitoring business.

[0083] Constructing a topological relationship graph , where the vertex set Each vertex corresponds to an advertising spot. Set spatial proximity threshold If the location with point The spherical distance between Less than or equal to the spatial proximity threshold Then in the topological relationship diagram of dense points vertex ,vertex Generate an undirected edge between them and mark the points. with point Points that are spatially close neighbors are considered to be adjacent; otherwise, no edge is established between them, thus obtaining the edge set. .

[0084] The LBS positioning and acquisition module collects multi-source status data from the monitoring terminal, including raw sensing data, monitoring personnel profile parameters, and traffic environment data around the location.

[0085] The system subscribes to the data streams reported by each online monitoring terminal and periodically collects the raw sensing data from each monitoring terminal, including the terminal's raw GNSS positioning output: longitude. ,latitude (Angle system); converting GNSS latitude and longitude to angle system Convert to radians for subsequent distance calculations:

[0086] Satellite signal-to-noise ratio set , The signal-to-noise ratio (SNR) of each effective positioning satellite is calculated; statistics are performed on the set of satellite SNRs to calculate the average effective satellite SNR at the current moment. ; This represents the number of valid satellites participating in the positioning at the current moment;

[0087] IMU inertial motion trajectory original sequence Representing the past A sequence of time trajectory points from one moment to the current moment t;

[0088] raw IMU trajectory The latitude and longitude of each trajectory point are also converted from angle to radians; invalid trajectory points with abnormal signals or out-of-bounds coordinates are filtered out to obtain the cleaned valid motion trajectory. ;

[0089] Access the personnel profile database to obtain the personnel profile parameters bound to the monitoring terminal, including historical verification accuracy. The current number of work orders accepted (Loadraw), and the original skill matching coefficient. ;

[0090] Among them, historical verification accuracy rate The percentage of work orders with correct manual review results among all of this person's historical verification work orders is calculated using the following formula:

[0091] ;

[0092] In the formula, This is the number of work orders that have been verified as correct in the past. This represents the total number of work orders verified by this person throughout history, with a value range of [value range missing]. ;

[0093] Number of work orders accepted so far This represents the number of monitoring work orders that the person currently has not yet completed.

[0094] Original skill matching coefficient The acquisition process is as follows:

[0095] Extract the advertising type tags of the current work orders to be monitored. Read the set of job tags for the monitoring personnel and determine whether the set contains advertising type tags. If this tag is not included, set the original skill matching score. The representative did not have the qualifications for this business position;

[0096] If the tag is included, read the preset tag weight corresponding to that tag. ;

[0097] Get the historical accuracy rate of similar work orders for this person ;

[0098] Based on the corresponding preset tag weight Based on the historical accuracy rate of similar work orders for this person Calculate the original skill matching score The formula is:

[0099] ;

[0100] Matching score for original skills Perform max-min normalization, the formula is:

[0101] ;

[0102] In the formula, For all system personnel Count the maximum value. For all system personnel Calculate the minimum value; output the normalized skill matching coefficient. .

[0103] Call the external traffic service interface to obtain the real-time raw congestion index of the area surrounding the advertising location to be monitored. ; A higher value indicates more congested roads, and this affects the original congestion index. Perform reverse normalization to obtain the normalized congestion factor. The more severe the congestion, the lower the output value of this factor.

[0104] The preprocessed and normalized data is assembled into a multi-source status data packet for the monitoring terminal; and the multi-source status data packets of all online monitoring terminals are output to the intelligent dispatching module to calculate the work order adaptation score of each terminal.

[0105] The aggregation and grouping module aggregates and groups work orders for spatially adjacent points based on the topological relationships of dense points, and resolves conflicts for duplicate work orders.

[0106] Receiver dense point topology association diagram Spatiotemporal location model of each point Read the location ID and deployment time window for each location. Receive all multi-source status data packets preprocessed by the online monitoring terminals;

[0107] Retrieve all pending monitoring work orders from the business database and construct a set of pending work orders. Each pending work order is bound to a unique location identifier. ;

[0108] Traverse the set of pending work orders Based on the topological relationship diagram Find the set of points that are spatially close neighbors;

[0109] For spatially adjacent points, read the deployment time window from their respective spatiotemporal point models. Determine if there is time overlap within the time window;

[0110] If the two points are in a time window , satisfy If so, then the time window overlap is satisfied;

[0111] Multiple work orders that simultaneously meet the criteria of spatial proximity and overlapping delivery time windows will be grouped into a single aggregated work order group. ,in For the m-th aggregation work order group, For monitoring work orders to be processed, neighboring locations that do not meet the time overlap requirement are retained as independent work orders and not aggregated.

[0112] It should be noted that although the spatial distance is close, the monitoring time windows do not overlap and are not aggregated to avoid forcibly packaging monitoring tasks from different time periods to the same monitoring terminal.

[0113] Iterate through aggregated work order groups and individual work orders to detect the same ad placement. Are there multiple unexecuted pending work orders at the same time?

[0114] If there are multiple duplicate work orders pending processing at the same location: read the business priority of each duplicate work order;

[0115] Only the work order with the highest business priority is retained as a valid work order; the remaining duplicate work orders are marked as redundant work orders, their status is set to archived and invalid, and they will not participate in subsequent work order dispatch; the aggregated work order group and the set of work orders to be processed are updated, and the invalidated redundant work orders are removed to obtain the set of valid work orders after conflict resolution.

[0116] The intelligent dispatching module calculates the adaptation score of each monitoring terminal based on multi-source status data for the locations within the current aggregated work order group, and issues monitoring work orders based on the adaptation score.

[0117] Obtain the reference latitude and longitude of all points within the aggregated work order group, and calculate the average latitude and longitude within the aggregated work order group based on the reference latitude and longitude of all points, which will serve as the spatial reference center point of the aggregated group. ;

[0118] Traverse all online monitoring terminals The current position of the monitoring terminal is calculated using the Haversine spherical distance formula. To the center point of the aggregation group spherical distance If satisfied The online monitoring terminal was included in the first-level candidate pool. ; Configurable distance filtering threshold;

[0119] Monitoring personnel not included in the above scope will be placed in the second-tier candidate pool. ;

[0120] Based on the current location of the monitoring terminal To the center point of the aggregation group spherical distance The distance factor is obtained by using reverse maximum-min normalization:

[0121] Obtain the number of monitoring work orders that are currently in transit and have not yet been completed by the monitoring personnel, and perform reverse max-min normalization on the number of monitoring work orders in transit to obtain the normalized load factor.

[0122] The normalized skill matching coefficient and congestion factor are obtained from multi-source state data, along with historical verification accuracy. The matching score is then calculated by weighting the normalized load factor and distance factor, as shown in the following formula:

[0123] ;

[0124] In the formula, For monitoring terminals Based on the overall compatibility score of the current work order group, the higher the score, the more suitable the monitoring terminal is to handle the work order group; , , , , The weighting coefficients for distance factor, congestion factor, historical verification accuracy, load factor, and skill matching coefficient are respectively, satisfying the constraints. The weights are configurable parameters of the system; This is the distance normalization factor; the closer the distance, the higher the factor value. To normalize the congestion factor, the higher the congestion level, the lower the factor value. The same congestion factor is shared by all monitoring terminals within the same aggregated work order group. To monitor the historical verification accuracy of terminals and characterize the quality of historical operations; To normalize the load factor, the more work orders in transit, the lower the factor value. This is a normalized skill matching coefficient; the higher the value, the better the terminal skill matches the current work order type.

[0125] If the set of valid monitoring personnel in the first-level candidate pool is not empty, all monitoring personnel in the first-level candidate pool are sorted in descending order according to their comprehensive suitability score, and a monitoring work order is issued to the monitoring personnel with the highest suitability score; if the monitoring personnel does not accept the issued work order, work orders are issued sequentially according to the comprehensive suitability score; if none of the valid monitoring personnel in the first-level candidate pool accept the issued work order or the set of valid monitoring personnel in the first-level candidate pool is empty, the second-level candidate pool is activated. Sort all monitoring personnel in the second-level candidate pool in descending order of their comprehensive fit score, and issue a monitoring work order to the monitoring personnel with the highest fit score in the second-level candidate pool; if the monitoring personnel does not accept the issued work order, issue work orders in order of comprehensive fit score.

[0126] The on-site verification module receives advertising verification multimedia materials submitted by the monitoring terminal, and performs on-site verification by combining the terminal's reported location, satellite signal-to-noise ratio, and IMU motion trajectory. Based on the verification results, it performs material interception, evidence storage, work order closure, and generates rectification and review work orders.

[0127] The system receives multimedia advertising verification materials reported by the monitoring terminal, and simultaneously obtains the terminal's GNSS positioning, satellite signal-to-noise ratio, and IMU motion trajectory at the time the materials were submitted. The multimedia materials are then bound to the location, trajectory, and signal-to-noise ratio data at the same time to obtain the associated verification dataset.

[0128] Read the GPS error compensation parameters of the advertising location, which include the historical average positioning error and the historical standard deviation of the positioning error. Calculate the basic fence radius of the location based on the 3σ principle.

[0129] The average satellite signal-to-noise ratio (SNR) for the time period during which the terminal submitted the verification materials is obtained. The SNR reflects the quality of the satellite signal; the better the signal, the higher the value, and the worse the signal, the lower the value. The average satellite SNR is normalized according to the upper and lower limits of the SNR preset by the system to obtain a normalized SNR value between 0 and 1.

[0130] The scaling factor generated by the normalized signal-to-noise ratio is used to correct the basic fence radius of the point location, so as to obtain the dynamic geofence radius. The upper and lower limits of the calculated dynamic fence radius are set to prevent the radius from being too large or too small and losing its verification significance.

[0131] Using the base geographical location of the advertising location as the center, a circle is drawn with the calculated and limited radius of the dynamic fence to form a circular dynamic geofence, which is used to determine whether the terminal has reached the location area.

[0132] Based on the IMU motion trajectory, a comprehensive judgment is made as to whether the monitoring terminal has actually arrived at the dynamic geofence area corresponding to the location; if it is determined that there is location spoofing or location drift, the multimedia material is intercepted for verification, and the work order remains in the pending verification state; after the location verification is passed, the material verification process begins.

[0133] Multi-modal verification of multimedia materials is carried out. The content of the on-site multimedia materials is compared with the advertising benchmark materials, and the on-site background environment is also compared to identify material tampering anomalies. The shooting timestamp is extracted from the EXIF ​​of the image and compared with the advertising placement time window. The verification confidence level is output by combining the location verification results, the degree of matching of advertising materials, the consistency of on-site background, and the matching of shooting time.

[0134] The system compares the verification confidence level with the preset verification pass threshold. When the verification confidence level is greater than or equal to the verification pass threshold, the verification is deemed to have passed. The system packages the location data, motion trajectory, multimedia materials, and timestamp into an evidence storage data package for evidence storage. The system pushes a verification completion message to the client to complete the work order closure. Work orders that need to be reviewed by the system are transferred to the manual review queue.

[0135] When the verification confidence level is less than the verification pass threshold, the verification is deemed unsuccessful, and a rectification and review work order is automatically generated and sent to the monitoring terminal. After the monitoring terminal resubmits the verification materials, all verification steps of this module are re-executed. After the second verification is completed, the work order is closed based on the verification results.

[0136] The above formulas are all dimensionless calculations, derived from software modeling based on collected large amounts of data.

[0137] A formula is proposed to obtain the most recent real situation, and the preset parameters in the formula are set by those skilled in the art according to the actual situation.

[0138] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0139] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0140] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0141] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0142] In conclusion, 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, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An outdoor advertising delivery intelligent dispatching monitoring platform based on LBS positioning, characterized in that, include: The advertising order access module acquires outdoor advertising placement orders, extracts advertising location information, constructs a spatiotemporal location model of advertising locations, and establishes topological relationships between dense locations. The LBS positioning and acquisition module collects multi-source status data from the monitoring terminal, including raw sensing data, monitoring personnel profile parameters, and traffic environment data around the location. The aggregation and grouping module aggregates and groups work orders for spatially adjacent points based on the topological relationships of dense points, and resolves conflicts for duplicate work orders. The intelligent dispatching module calculates the adaptation score of each monitoring terminal based on multi-source status data for the locations within the current aggregated work order group, and issues monitoring work orders based on the adaptation score. The on-site verification module receives advertising verification multimedia materials submitted by the monitoring terminal, and performs on-site verification by combining the terminal's reported location, satellite signal-to-noise ratio, and IMU motion trajectory. Based on the verification results, it performs material interception, evidence storage, work order closure, and generates rectification and review work orders. 2.The LBS positioning-based outdoor advertising delivery intelligent dispatching monitoring platform of claim 1, wherein, The process of constructing the spatiotemporal location model of advertising positions is as follows: Receive outdoor advertising placement order messages and verify the legality of the order messages; After successful verification, extract the advertising location information, including the unique identifier of the location, the reference latitude and longitude of the location, the advertising type, the start time of the advertising campaign, the end time of the advertising campaign, and the reference material sample set for the advertising campaign. Read the historical GNSS positioning statistics log of this point, calculate the positioning error of multiple historical positioning results of this point relative to the reference position, calculate the mean and standard deviation of historical positioning error, and combine them to generate GPS error compensation parameters for this point. Based on the GPS error compensation parameters of this location, combined with the unique identifier of the location, the reference latitude and longitude of the location, the start time of the advertising campaign, the end time of the advertising campaign, and the reference material sample set of the advertising campaign, a spatiotemporal location model of this location is assembled. 3.The LBS positioning-based outdoor advertisement delivery intelligent dispatching monitoring platform of claim 2, wherein, The process of establishing topological relationships between densely populated points is as follows: Iterate through the entire set of existing ad locations and use the Haversine formula to calculate the spherical geodetic distance between location i and location j. Construct a dense topological graph of ad placements, where each vertex in the vertex set corresponds to an ad placement. Set a spatial proximity threshold. If the spherical distance between point i and point j is less than or equal to the spatial proximity threshold, then generate an undirected connection edge between vertex i and vertex j in the dense point topology graph, marking point i and point j as spatially dense neighbors. Otherwise, do not establish a connection edge between them, thus obtaining the edge set. 4.The LBS positioning-based outdoor advertisement delivery intelligent dispatching monitoring platform of claim 3, wherein, The raw sensing data includes terminal GNSS positioning, satellite signal-to-noise ratio, and effective motion trajectory. 5.The LBS positioning-based outdoor advertising delivery intelligent dispatching monitoring platform of claim 4, wherein, The parameters for monitoring personnel profiles include historical verification accuracy, the number of work orders currently accepted, and the original skill matching coefficient. The historical verification accuracy rate is the percentage of work orders with correct manual review results among all the personnel's historical verification work orders. The number of work orders currently accepted is the number of monitoring work orders that the person has not yet completed and are currently in transit. Original skill matching coefficient The acquisition process is as follows: Extract the advertising type tag of the current work order to be monitored; read the job tag set of the monitored personnel and determine whether the set contains the advertising type tag; if the tag is not contained, set the original skill matching score to 0, which means that the personnel do not have the qualifications for the job. If the tag is included, read the preset tag weight corresponding to that tag; Obtain the historical accuracy rate of similar work orders for this person; The original skill matching score is calculated based on the corresponding preset label weights and the historical accuracy rate of the person's similar types of work orders. Perform maximum-min normalization on the original skill matching score and output the normalized skill matching coefficient. 6.The LBS positioning-based outdoor advertising delivery intelligent dispatching monitoring platform of claim 5, wherein, The process of aggregating and grouping work orders for spatially adjacent points based on the topological relationships of densely populated points is as follows: Receive the topological relationship map of dense points and the spatiotemporal point model of each point, and read the point ID and deployment time window of each point; Receive all multi-source status data packets preprocessed by the online monitoring terminals; All pending monitoring work orders are retrieved from the business database, and a set of pending work orders is constructed. Each pending work order is bound to a unique location identifier. Traverse the set of pending work orders and find the set of points that are spatially close to each other based on the topological relationship graph; For spatially adjacent points, read the deployment time window in their respective spatiotemporal point models and determine whether the time windows overlap. Work orders that simultaneously meet the criteria of spatial proximity and overlapping delivery time windows will be grouped into the same aggregated work order group.

7. The LBS-based intelligent order dispatching and monitoring platform for outdoor advertising as described in claim 6, characterized in that, The conflict resolution process for duplicate work orders is as follows: If there are duplicate work orders pending at the same location, read the business priority of each duplicate work order; Only the work order with the highest business priority will be retained as a valid work order; Other duplicate work orders are marked as redundant work orders, their status is set to archived and voided, and they will not participate in subsequent work order dispatch. Update the aggregated work order group and the set of work orders to be processed, remove redundant work orders that have been invalidated, and obtain the set of valid work orders after conflict resolution.

8. The LBS-based intelligent order dispatch monitoring platform for outdoor advertising as described in claim 7, characterized in that, The process of calculating the adaptation score of each monitoring terminal based on multi-source status data is as follows: Obtain the reference latitude and longitude of all points within the aggregated work order group, and calculate the average latitude and longitude within the aggregated work order group based on the reference latitude and longitude of all points, which serves as the spatial reference center point of the aggregated group. Traverse all online monitoring terminals and use the Haversine spherical distance formula to calculate the spherical distance from the current position of the monitoring terminal to the center point of the aggregation group; Based on the spherical distance from the current location of the monitoring terminal to the center point of the aggregation group, the distance factor is obtained using reverse maximum-minimum normalization: Obtain the number of monitoring work orders that are currently in transit and have not yet been completed by the monitoring personnel, and perform reverse max-min normalization on the number of monitoring work orders in transit to obtain the normalized load factor. The normalized skill matching coefficient and congestion factor are obtained from multi-source state data, and the historical verification accuracy is calculated. The matching score is obtained by weighting the normalized load factor and distance factor.

9. The LBS-based intelligent order dispatch monitoring platform for outdoor advertising as described in claim 8, characterized in that, The process of issuing monitoring work orders based on the adaptation score is as follows: If the spherical distance from the current location of the monitoring terminal to the center point of the aggregation group is less than or equal to the configurable distance filtering threshold, the online monitoring terminal will be included in the first-level candidate pool. If the spherical distance from the current location of the monitoring terminal to the center point of the aggregation group is greater than the configurable distance filtering threshold, the monitoring terminal will be classified into the second-level candidate pool. If the set of valid monitoring personnel in the first-level candidate pool is not empty, then all monitoring personnel in the first-level candidate pool are sorted in descending order according to their comprehensive suitability scores, and a monitoring work order is issued to the monitoring personnel with the highest suitability score; if the monitoring personnel does not accept the issued work order, then work orders are issued in order according to the comprehensive suitability scores. If none of the valid monitoring personnel in the first-level candidate pool have accepted the issued work order or the set of valid monitoring personnel in the first-level candidate pool is empty, then the second-level candidate pool will be activated. All monitoring personnel in the second-level candidate pool are sorted in descending order according to their comprehensive fit score. A monitoring work order is issued to the monitoring personnel with the highest fit score in the second-level candidate pool. If the monitoring personnel does not accept the issued work order, work orders are issued in order according to the comprehensive fit score.

10. The LBS-based intelligent order dispatching and monitoring platform for outdoor advertising as described in claim 9, characterized in that, The specific implementation process of the on-site verification module is as follows: The system receives advertising verification multimedia materials reported by the monitoring terminal, and binds the multimedia materials with the location, trajectory, and signal-to-noise ratio data at the same time to obtain the associated verification dataset. Read the GPS error compensation parameters of the advertising location and calculate the basic fence radius of the location according to the 3σ principle; Obtain the average satellite signal-to-noise ratio (SNR), and normalize the average satellite SNR according to the preset upper and lower limits of SNR to obtain the normalized SNR value. The normalized signal-to-noise ratio is used to generate a scaling factor to correct the basic fence radius of the point location, thus obtaining the dynamic geofence radius; Using the base geographical location of the advertising location as the center, draw a circle with the radius of the dynamic fence to form a dynamic geofence; By combining the IMU motion trajectory, a comprehensive judgment is made as to whether the monitoring terminal has actually arrived at the dynamic geofence area corresponding to the location. If location spoofing or location drift is detected, the multimedia material will be intercepted for verification, and the work order will remain in the pending verification state. After the location verification is passed, the material verification process begins; Based on the comprehensive location verification results, the matching degree of advertising materials, the consistency of the on-site background, and the matching of shooting time, the verification confidence level is output. The verification confidence level is compared with the preset verification. When the verification confidence level is greater than or equal to the verification pass threshold, the verification is determined to be passed. The location data, motion trajectory, multimedia materials, and timestamp are packaged into a data package for evidence storage. The system marks the work order that needs to be reviewed and transfers it to the manual review queue. When the verification confidence level is less than the verification pass threshold, the verification is deemed unsuccessful, and a rectification and review work order is generated and sent to the monitoring terminal. After the monitoring terminal resubmits the verification materials, the verification steps are executed again. After the second verification is completed, the work order is closed based on the verification results.