A running monitoring method, device, equipment and storage medium

Abstract

The present application relates to the technical field of running monitoring, and particularly relates to a running monitoring method, device, equipment and storage medium; each management object is configured according to a preset campus tree level and a virtual fence configuration parameter, so as to obtain a plurality of configuration management objects; a running test instruction is generated according to each configuration management object and a virtual fence management area, and a running test state is entered according to the running test instruction; in the running test state, positioning position data and motion sensor data are analyzed according to a preset cheating risk assessment method and a preset multi-level anti-cheating analysis mechanism, so as to obtain a verification analysis result; a running monitoring result is generated according to the positioning position data, the motion sensor data and the verification analysis result; by building a campus running whole-process intelligent monitoring system, relying on a campus level architecture and a virtual fence to realize differentiated scene adaptation, combining a multi-level progressive anti-cheating verification mechanism to identify running abnormalities, and guaranteeing the standard fairness of campus sports examination.

Description

Technical Field

[0001] This invention relates to the field of running monitoring technology, and in particular to a running monitoring method, device, equipment, and storage medium. Background Technology

[0002] Current campus intelligent running monitoring technologies generally adopt unified monitoring rules and a single GPS positioning verification mode, resulting in a crude overall control approach and significant shortcomings in adaptability and anti-cheating capabilities. Existing technologies cannot adapt to diverse campus running assessment scenarios, lack a multi-source sensor data linkage verification mechanism for movement posture, stride frequency, stride length, etc., making it difficult to effectively identify cheating behaviors such as virtual positioning, commuter cycling, and trajectory simulation, resulting in low detection accuracy. In addition, the monitoring rules of the front-end and back-end of existing technologies are prone to inconsistencies, lacking command control and full-process trajectory traceability and playback capabilities, making it difficult to accurately review and verify abnormal running behaviors, leading to misjudgments and omissions. This results in insufficient fairness and standardization in campus running assessments, failing to meet the needs of modern campus sports for refined, intelligent, and traceable running monitoring management. Summary of the Invention

[0003] In order to overcome the shortcomings of the prior art, the purpose of this invention is to provide a running monitoring method, device, equipment and storage medium.

[0004] A running monitoring method includes: acquiring virtual fence configuration parameters, multiple managed objects, and virtual fence management areas of a virtual fence system; configuring each managed object according to a preset campus tree hierarchy and virtual fence configuration parameters to obtain multiple configured managed objects; generating running test instructions based on each configured managed object and the virtual fence management area, and entering a running test state according to the running test instructions; acquiring location data and motion sensor data of each configured managed object in the running test state; analyzing the location data and motion sensor data according to a preset cheating risk assessment method and a preset multi-level anti-cheating analysis mechanism to obtain verification analysis results; and generating running monitoring results based on the location data, motion sensor data, and verification analysis results.

[0005] Furthermore, the step of generating running test instructions based on each configuration management object and the virtual fence management area includes: identifying and analyzing each configuration management object according to a preset quadtree spatial retrieval method to obtain coverage identification results and multiple computational complexities; adjusting the coverage identification results according to a preset spatial hash collision coarse screening method, various computational complexities, preset first initial screening conditions, and preset second initial screening conditions to obtain adjusted identification results; and generating running test instructions based on the adjusted identification results and the virtual fence management area.

[0006] Furthermore, the step of identifying and analyzing each configuration management object according to a preset quadtree spatial retrieval method to obtain coverage identification results and multiple computational complexities includes: obtaining the geographical coordinates of each configuration management object to obtain a set of coordinate points; performing geometric construction on the set of coordinate points according to a preset polygon intersection detection method to obtain multiple fence polygons; and identifying and analyzing each fence polygon according to the quadtree spatial retrieval method and a preset geometric length threshold to obtain coverage identification results and multiple computational complexities.

[0007] Furthermore, the adjustment of the coverage identification result based on the preset spatial hash collision coarse screening method, various computational complexities, preset first initial screening conditions, and preset second initial screening conditions to obtain the adjusted identification result includes: encoding the coordinate point set according to the spatial hash collision coarse screening method and a preset global mapping dictionary to obtain multiple latitude and longitude codes; filtering each computational complexity according to the first initial screening conditions and preset complexity thresholds to obtain the first initial screening result; and adjusting the coverage identification result according to the second initial screening conditions, the first initial screening result, and each latitude and longitude code to obtain the adjusted identification result.

[0008] Furthermore, the step of adjusting the coverage identification result based on the second initial screening conditions, the first initial screening result, and each latitude and longitude code to obtain the adjusted identification result includes: screening each latitude and longitude code based on the second initial screening conditions and a preset latitude code threshold to obtain the second initial screening result; and adjusting the coverage identification result based on the first initial screening result and the second initial screening result to obtain the adjusted identification result.

[0009] Furthermore, the step of generating running test instructions based on the adjusted recognition results and the virtual fence management area includes: analyzing each fence polygon according to preset fence extraction conditions, adjusted recognition results, and preset deconstruction model algorithm to obtain a set of non-covered fences; configuring the non-covered fence set according to preset running rule packages and each configuration management object to obtain a set of configured fences; dividing the virtual fence management area according to the set of coordinate points to obtain multiple management areas; and generating running test instructions based on the set of covered fences and each management area.

[0010] Furthermore, the step of analyzing the location data and motion sensor data according to the preset cheating risk assessment method and the preset multi-level anti-cheating analysis mechanism to obtain the verification analysis result includes: evaluating and analyzing the location data and motion sensor data according to the cheating risk assessment method and the coverage fence set to obtain the risk assessment result; and verifying the risk assessment result according to the multi-level anti-cheating analysis mechanism, the preset time interval, and the preset distance interval to obtain the verification analysis result.

[0011] Furthermore, a running monitoring device includes: a first data acquisition module for acquiring virtual fence configuration parameters, multiple managed objects, and virtual fence management areas of a virtual fence system; a configuration module for configuring each managed object according to a preset campus tree hierarchy and virtual fence configuration parameters to obtain multiple configured managed objects; a testing module for generating running test instructions based on each configured managed object and the virtual fence management area, and entering a running test state according to the running test instructions; a second data acquisition module for acquiring location data and motion sensor data of each configured managed object during the running test state; an analysis module for analyzing the location data and motion sensor data according to a preset cheating risk assessment method and a preset multi-level anti-cheating analysis mechanism to obtain verification analysis results; and a result generation module for generating running monitoring results based on the location data, motion sensor data, and verification analysis results.

[0012] Furthermore, a running monitoring device includes: a memory and at least one processor, the memory storing instructions; at least one processor invokes the instructions in the memory to cause the running monitoring device to perform the steps of the running monitoring method as described above.

[0013] Furthermore, a computer-readable storage medium stores instructions that, when executed by a processor, implement the steps of the running monitoring method as described above.

[0014] The technical solution of this invention constructs a full-process monitoring system for campus running, encompassing parameter configuration, command issuance, data collection, hierarchical verification, and result output. It leverages a hierarchical campus tree structure combined with virtual fence parameters to configure various managed objects, adapting to diverse campus running management scenarios. Running test commands unify front-end and back-end monitoring rules, ensuring consistency between terminal data collection and back-end judgment logic. Simultaneously, it integrates GPS positioning data and multi-source motion sensor time-series data to fully reconstruct the spatiotemporal movement state of students during running. Coupled with a progressive multi-level anti-cheating verification mechanism, it identifies various cheating behaviors from multiple dimensions, including basic indicators, physiological correlations, and trajectory consistency, and hierarchically marks abnormal movement risks. The entire solution achieves refined, differentiated, and intelligent management of campus running, improving the accuracy of running authenticity verification and ensuring the fairness, standardization, and credibility of campus sports assessments. Attached Figure Description

[0015] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, in which: Figure 1 This is a first flowchart of a running monitoring method provided in an embodiment of the present invention; Figure 2 This is a second flowchart of a running monitoring method provided in an embodiment of the present invention; Figure 3 This is a third flowchart of a running monitoring method provided in an embodiment of the present invention; Figure 4 This is a fourth flowchart of a running monitoring method provided in an embodiment of the present invention; Figure 5 This is a fifth flowchart of a running monitoring method provided in an embodiment of the present invention; Figure 6 A sixth flowchart of a running monitoring method provided in an embodiment of the present invention; Figure 7 A seventh flowchart of a running monitoring method provided in an embodiment of the present invention; Figure 8 This is a schematic diagram of the structure of a running monitoring device provided in an embodiment of the present invention; Figure 9 This is a schematic diagram of the structure of a running monitoring device provided in an embodiment of the present invention. Detailed Implementation

[0016] The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0017] For ease of understanding, the specific process of the embodiments of the present invention is described below. Please refer to [link / reference]. Figure 1 One embodiment of a running monitoring method according to the present invention includes: 101. Obtain the virtual fence configuration parameters, multiple managed objects, and virtual fence management area of ​​the virtual fence system; In this embodiment, the system uniformly reads the preset configuration information within the virtual fence system, and summarizes and obtains the virtual fence configuration parameters, all managed object information, and virtual fence management area data. The virtual fence configuration parameters support differentiated definitions based on multiple attributes such as grade, college, class, and individual, and carry differentiated fence boundaries, opening time rules, running operation modes, and cheating detection level control rules. The managed objects correspond to the main students participating in running monitoring on campus. The virtual fence management area delineates the geographical range within which each group is allowed to conduct running activities. By uniformly collecting the three types of basic data, the system completes the collection of the original business parameters required for the campus running monitoring task, providing complete data source support for subsequent hierarchical object configuration. 102. Configure each management object according to the preset campus tree hierarchy and virtual fence configuration parameters to obtain multiple configuration management objects; In this embodiment, a pre-defined campus tree hierarchy is used to build a top-down hierarchical management structure, covering multiple levels of affiliation, including colleges, grades, classes, and individuals. The system categorizes managed objects based on this tree hierarchy and matches corresponding virtual fence configuration parameters, binding exclusive fence boundaries, open times, running modes, and anti-cheating levels to different levels and student objects. After parameter binding and attribute encapsulation, multiple configuration management objects are generated. This enables differentiated rule allocation based on the campus organizational structure, meeting the diverse running management needs of different groups and providing standardized management object units for running task initialization. 103. Generate running test instructions based on each configuration management object and virtual fence management area, and enter the running test state according to the running test instructions; In this embodiment, the system associates the permission parameters corresponding to each configuration management object with the geographical range of the virtual fence management area, integrates fence boundary constraints, running mode rules, check-in requirements, and anti-cheating strategies, and uniformly encapsulates them into running test instructions. The system sends running test instructions to student terminals, the terminals parse the instructions to complete the initialization of monitoring parameters, and automatically switch to the running test state. The running test instructions uniformly standardize the front-end and back-end judgment criteria, clarify all constraints of this running task, ensure that the terminal collection logic and the back-end monitoring rules are consistent, and provide an execution basis for real-time sports data collection and boundary discrimination. 104. During the running test, obtain the location data and motion sensor data of each configuration management object; In this embodiment, the terminal continuously executes the data acquisition process, synchronously collecting GPS positioning data and motion sensor data composed of accelerometers and gyroscopes according to a preset sampling period. The positioning data reflects the student's geographical coordinates in real time, while the motion sensor data represents movement characteristics such as running posture, cadence, and number of steps. Both types of data are continuously uploaded to the background and form a time-series data stream, completely recording the spatiotemporal information and movement status of the student throughout the entire running process. This provides multi-source raw monitoring data support for fence crossing identification, movement authenticity verification, and segmented risk assessment. 105. Analyze the positioning data and motion sensor data according to the preset cheating risk assessment method and the preset multi-level anti-cheating analysis mechanism to obtain the verification analysis results; In this embodiment, a progressive and refined verification is implemented based on a multi-level anti-cheating analysis mechanism. The system divides the complete running process into multiple exercise segments according to time intervals or distance intervals, and sequentially performs basic indicator verification, stride frequency and stride physiological correlation verification, and GPS trajectory spatiotemporal consistency verification, distinguishing between low, medium, and high detection intensities. It comprehensively identifies various cheating behaviors such as virtual positioning, using vehicles, trajectory drift, and abnormal displacement, and marks suspected abnormal segments. The segment verification conclusions are integrated to form a verification analysis result, distinguishing between normal exercise, suspected abnormality, and high cheating risk levels, providing a tiered risk basis for the final score determination. 106. Generate running monitoring results based on location data, motion sensor data, and verification analysis results; In this embodiment, the system integrates time-series location data, motion sensor data, and verification analysis results from the preceding stages to comprehensively generate running monitoring results. During the running process, the student terminal continuously collects GPS location points, timestamps, instantaneous speed, real-time pace, steps, cadence, checkpoint status, and fence entry / exit status, and simultaneously associates anti-cheating segment detection information. The system constructs a complete motion trajectory sequence according to the time series, and marks the running status for each trajectory point and motion segment, including normal motion segments, boundary crossing pause segments, checkpoint checkpoint segments, pace abnormal segments, and high-risk abnormal segments. After the run, the terminal uploads the full trajectory data for service. The system compresses, encodes, segments, indexes, and persistently stores the trajectory data. This data is used for automated performance evaluation and supports dynamic trajectory replay for the management backend and student clients. During replay, the system fully reconstructs the entire running process along the timeline, displaying the start point, finish line, route, check-in completion status, boundary crossing pauses, pace fluctuations, and various abnormal segment markers. Administrators use the replay information to verify whether students have completed the running task as required, identifying violations such as location jumps, trajectory drift, abnormal high-speed displacement, avoidance of check-in points, and continuous boundary crossings. Students can view their own running trajectory and the reasons for invalidated scores, providing support for appeals and reviews. Intuitive evidence; the dynamic playback and performance evaluation of the running track is not merely a simple display of exercise records, but rather relies on multi-source monitoring data to achieve structured storage, visual reproduction, and rule-based automatic evaluation of the entire running process, forming a complete chain of evidence and an automated scoring mechanism to support campus running management; during the performance evaluation stage, the system calls the rule engine to conduct multi-dimensional automatic verification, verifying whether all required checkpoints have been completed, whether the single valid mileage has met the standard, whether the overall and segmented paces fall within a reasonable range, whether there are any non-compliant interruptions, and whether there are any high-risk cheating markers; differentiated verification logic is executed for different running modes: the checkpoint running mode focuses on... The system verifies the completion of check-in points and the order of movement; in the no-check-in mode, it focuses on verifying the fence coverage, effective mileage, and pace indicators; in the high-level anti-cheating mode, it additionally verifies the number of abnormal segments, the percentage of abnormal segments, and high-risk markers; the system uses a multi-condition joint judgment mechanism to output the final result; when the check-in task is completed, the mileage is met, the pace is compliant, and there are no high-risk cheating markers, the run is deemed valid and included in the student's running record, cumulative mileage, valid attempts, and leaderboard; if the key verification conditions are not met, the running record is deemed invalid or included in the manual review queue, and a detailed explanation of the reason is generated simultaneously, and finally all information is integrated to form the running monitoring result; In this embodiment, a comprehensive campus running monitoring system was constructed, encompassing parameter configuration, command issuance, data collection, hierarchical verification, and result output. Utilizing a hierarchical campus tree structure combined with virtual fence parameters, the system enables configuration of various managed objects, adapting to diverse campus running management scenarios. Running test commands unify front-end and back-end monitoring rules, ensuring consistency between terminal data collection and back-end judgment logic. Simultaneously, GPS location data and multi-source motion sensor time-series data are integrated to fully reconstruct the spatiotemporal movement state of students during running. Coupled with a progressive, multi-level anti-cheating verification mechanism, various cheating behaviors are identified from multiple dimensions, including basic indicators, physiological correlations, and trajectory consistency, with hierarchical marking of abnormal movement risks. This entire solution achieves refined, differentiated, and intelligent management of campus running, improving the accuracy of running authenticity verification and ensuring the fairness, standardization, and credibility of campus physical education assessments.

[0018] Please see Figure 2 A second embodiment of a running monitoring method according to the present invention includes: 201. Based on the preset quadtree space retrieval method, identify and analyze each configuration management object to obtain the coverage identification results and multiple computational complexities; In this embodiment, a pre-defined quadtree spatial retrieval method is used to build a hierarchical spatial query framework for spatial matching and identification of all configuration management objects. The system relies on the geographic coordinates of the configuration management objects and uses a quadtree recursive spatial partitioning mechanism to progressively narrow the comparison range, quickly determining the spatial coverage relationship between the configuration management objects and the fenced area. Spatial matching operations are completed by traversing each configuration management object, and the coverage identification results are output. Simultaneously, the computational resources required for each spatial matching task are evaluated by combining the number of fence polygon vertices, spatial layer depth, and intersection operation scale, generating multiple corresponding computational complexities. The coverage identification results represent the preliminary conclusions of spatial matching, and the computational complexity measures the computational power consumption of each task, providing two types of input parameters to support subsequent multi-condition joint screening. 202. Adjust the coverage recognition results according to the preset spatial hash collision coarse screening method, various computational complexities, preset first initial screening conditions and preset second initial screening conditions to obtain the adjusted recognition results; In this embodiment, a pre-defined spatial hash collision coarse screening method is used to discretely group geographic points. The first initial screening condition is based on computational complexity to construct a computing power admission rule, and the second initial screening condition sets geographic interval screening constraints. The system integrates the spatial grouping information output by the hash coarse screening, various computational complexity indicators, and two-level initial screening judgment rules to comprehensively correct the original coverage identification results obtained by quadtree retrieval. Filtering is carried out sequentially from two dimensions: computing power load and geographic location validity. False coverage matching entries caused by computing power exceeding limits or points being outside the effective monitoring range are eliminated. Valid identification entries are sorted, corrected, and summarized. The mismatch defects caused by single quadtree retrieval are eliminated, and optimized adjusted identification results are obtained, providing a reliable spatial judgment basis for the generation of running test instructions. 203. Generate running test instructions based on the adjusted recognition results and the virtual fence management area; In this embodiment, the adjusted recognition results carry spatial matching conclusions after multi-layer screening and optimization, and the virtual fence management area delineates the overall geographic control range corresponding to the campus running business; the system associates and binds the effective spatial matching relationships within the adjusted recognition results with each virtual fence management area, matching the corresponding configuration management object's belonging area, effective sports fence range, preset running rules, and anti-cheating parameters; all constraint parameters such as fence boundaries, assessment indicators, check-in strategies, and anti-cheating levels are uniformly encapsulated, generating running test instructions and sending them to student terminals; the running test instructions fully carry all control rules for this running task, ensuring that the backend monitoring logic and terminal execution standards remain consistent, supporting the orderly execution of subsequent location collection, fence boundary discrimination, sports data monitoring, and layered anti-cheating verification; In this embodiment, quadtree spatial retrieval is used to complete the initial spatial identification of the configuration management object, and the coverage identification results and quantified computational complexity are output simultaneously, so as to achieve simultaneous acquisition of spatial matching results and computing power consumption. Based on the spatial hash collision coarse screening method and two-level initial screening conditions, the original identification results are corrected and filtered from both computing power load and geospatial validity perspectives, effectively eliminating invalid matching data and improving the accuracy of spatial judgment. Furthermore, the running test instructions are uniformly packaged with the adjusted identification results and virtual fence management area, so as to achieve simultaneous distribution of fence range, running rules and anti-cheating strategies, ensuring that the judgment standards of the terminal and the backend are consistent. The whole process takes into account the computational efficiency and spatial identification accuracy in multi-user concurrent scenarios, providing a stable and unified spatial control foundation for the whole process monitoring of campus running.

[0019] Please see Figure 3 A third embodiment of a running monitoring method according to the present invention includes: 301. Obtain the geographic coordinates of each configuration management object to obtain a set of coordinate points; In this embodiment, the system collects the geographic coordinate information of all configuration management objects and summarizes it into a complete set of coordinate points. The set of coordinate points carries the original spatial location data of different management objects and records the geographic distribution information of each configuration management object, serving as the original data source for subsequent geometric construction and spatial retrieval analysis. By uniformly collecting geographic coordinate points, the system realizes the basic spatial data and summary output, providing original spatial parameter support for polygon geometric modeling and fence spatial recognition. 302. Based on the preset polygon intersection detection method, the coordinate point set is geometrically constructed to obtain multiple fence polygons; In this embodiment, spatial geometric modeling is carried out based on the collected set of coordinate points, and a preset polygon intersection detection method is used as the geometric operation benchmark. The system uses the geographic coordinates in the coordinate point set to sequentially complete the boundary connection and closed region construction according to the fence boundary point arrangement rules, generating closed fence polygons of the corresponding geographic range. The polygon intersection detection method predefines the logic of line segment intersection, boundary bounding box pre-judgment, and region belonging discrimination to ensure the accuracy of the boundary topology relationship during the geometric construction process. Through point information vectorization processing, discrete coordinate points are transformed into fence polygons with clear boundary ranges, resulting in multiple fence polygons. This realizes the transformation of discrete points into geometric regions that can participate in spatial operations, providing geometric objects for subsequent quadtree retrieval. 303. Based on the quadtree spatial retrieval method and the preset geometric length threshold, identify and analyze each fence polygon to obtain the coverage identification results and multiple computational complexities; In this embodiment, a quadtree spatial retrieval method is used to perform spatial range matching and identification on all fence polygons. Combined with a preset geometric length threshold, polygon hierarchical filtering is completed. The quadtree spatial retrieval method recursively divides the geographic space into multi-level subspaces, quickly narrowing down the range of geometric objects to be compared and reducing the overhead of global traversal. The preset geometric length threshold is used to define the size admission conditions for fence polygons to participate in spatial matching, filtering out polygons that are too small to meet business requirements. The system performs spatial coverage matching operations on each fence polygon that meets the conditions, determining the coverage relationship between the fence polygon and the spatial range of the configuration management object, and outputting the coverage identification results. Simultaneously, based on the number of polygon vertices, spatial level depth, and intersection calculation scale, the corresponding computational complexity is evaluated and generated in real time, representing the computational cost required for subsequent in-depth spatial analysis of each fence polygon. The coverage identification results and computational complexity are output synchronously, providing a dual-dimensional analysis basis for subsequent multi-level initial screening and fence correction processing. In this embodiment, the geographical coordinates of each configured management object are aggregated to form a coordinate point set, thus unifying and standardizing the original spatial data and providing a reliable data foundation for subsequent geometric operations. A polygon intersection detection method is used to vectorize discrete coordinates to construct fence polygons, restoring the topological structure of the campus virtual fence boundary and ensuring the accuracy of geographical area modeling. Combined with geometric length threshold filtering and a quadtree spatial retrieval mechanism, invalid small fence areas are filtered out. Spatial recursive partitioning reduces the comparison range and lowers the computational overhead of the entire domain retrieval. Simultaneously, the coverage recognition results and multi-dimensional computational complexity are output synchronously, achieving dual quantification of spatial matching effect and computational power consumption. This provides precise support for subsequent multi-level initial screening and fence result correction, effectively improving the efficiency and accuracy of virtual fence recognition in scenarios with multiple management objects, and ensuring the stability and standardization of campus running monitoring spatial control.

[0020] Please see Figure 4 A fourth embodiment of a running monitoring method according to the present invention includes: 401. Encode the set of coordinate points using the spatial hash collision coarse screening method and the preset global mapping dictionary to obtain multiple latitude and longitude codes; In this embodiment, the coordinate point set output by the preprocess carries the geographic location information of all configuration management objects. Geographic information encoding conversion is carried out by combining the spatial hash collision coarse screening method with the preset global mapping dictionary. The global mapping dictionary stores the one-to-one mapping relationship between latitude and longitude intervals and hash codes as the benchmark for geographic coordinate discretization conversion. The system calls the spatial hash collision coarse screening method to map the continuous latitude and longitude coordinates in the coordinate point set to discrete hash intervals, and completes the matching encoding with reference to the global mapping dictionary to generate multiple one-to-one latitude and longitude codes. Through the continuous geographic coordinate discretization processing, the geographic location is quickly classified. The natural grouping characteristics of hash are used to complete the coarse grouping of geographic points, providing standardized discrete feature parameters to support the subsequent multi-level initial screening. 402. Based on the first initial screening conditions and the preset complexity threshold, each computational complexity is screened to obtain the first initial screening result; In this embodiment, each computational complexity represents the scale of computing resources required to perform spatial intersection recognition on the corresponding fence polygon; a first preliminary screening condition is set up in combination with a fixed complexity threshold to construct a computing power admission screening rule, which is used to eliminate analysis objects with high computational overhead and unnecessary deep computation; the system compares each computational complexity with the preset complexity threshold one by one, filters out recognition objects whose computational load is within the threshold range, filters out targets that exceed the computing power capacity limit and have low spatial analysis value, and outputs the first preliminary screening result; through the computational complexity pre-filtering mechanism, the amount of spatial geometry computation tasks to be executed is reduced in advance, preventing the simultaneous calculation of a large number of complex polygons from causing system concurrency blocking, realizing lightweight preprocessing of computational tasks, and improving the overall spatial retrieval link operation efficiency; 403. Adjust the coverage identification results based on the second initial screening conditions, the first initial screening results, and the latitude and longitude codes to obtain the adjusted identification results; In this embodiment, a preset second initial screening condition constructs a spatial range verification logic, relying on the discretized latitude and longitude codes to determine the validity of geographical locations. The system simultaneously integrates the first initial screening result representing computing power constraints with the latitude and longitude code verification conclusion representing spatial constraints, and performs comprehensive correction on the original coverage identification results output by the quadtree retrieval. For each identification entry in the original coverage identification results, it simultaneously verifies whether it meets the computational complexity threshold requirements and the geographical interval validity requirements, and eliminates false matching entries that exceed the computing power limit or whose locations are outside the target monitoring area. Through bidirectional joint verification of computing power dimension and geographic space dimension, it corrects the mismatch problem caused by single quadtree spatial retrieval, completes the screening, elimination and sorting of identification entries, and obtains the adjusted identification results, providing a reliable spatial judgment basis for subsequent fence polygon parsing and running test command generation. In this embodiment, continuous latitude and longitude coordinates are discretely encoded by combining spatial hash collision coarse screening with a global mapping dictionary, enabling rapid grouping of geographic points and providing standardized data support for refined screening. Simultaneously, a first initial screening condition is used to match a complexity threshold to complete computational power-level filtering, preemptively eliminating high-load, invalid computation tasks and streamlining the spatial computation process. Furthermore, a second initial screening condition is combined to fuse computational power screening results with latitude and longitude spatial verification results, thus correcting the coverage and recognition results from a dual-dimensional perspective, effectively eliminating false spatial matching data and correcting the recognition bias of a single retrieval method. This solution combines computational power optimization with precise spatial verification, reducing the geometric computation pressure in concurrent scenarios involving multiple managed objects on campus, improving the operational efficiency of virtual fence spatial recognition, and providing efficient and stable spatial data support for subsequent fence parsing, rule configuration, and running monitoring task generation.

[0021] Please see Figure 5 The fifth embodiment of a running monitoring method in this invention includes: 501. Based on the second preliminary screening conditions and the preset latitude coding threshold, each latitude and longitude code is screened to obtain the second preliminary screening results; In this embodiment, the preset second initial screening conditions configure geographic interval filtering logic, and combine latitude coding threshold to construct latitude and longitude coding screening criteria, which are used to perform spatial dimension verification on all latitude and longitude codes; the judgment logic is to screen out valid coding entries whose latitude codes fall within the latitude coding threshold limit range, and remove latitude and longitude codes whose latitude range deviates significantly from the target monitoring area; by comparing and verifying each discretized latitude and longitude code, the secondary screening of the spatial range of geographic points is completed, and the second initial screening result is output, providing a spatial dimension judgment basis for subsequent fusion of multi-dimensional initial screening conclusions to correct the coverage identification results; 502. Adjust the coverage identification results based on the first and second preliminary screening results to obtain the adjusted identification results; In this embodiment, the first initial screening result carries the computational complexity level of the computational power screening conclusion, while the second initial screening result carries the spatial range screening conclusion based on latitude and longitude encoding and latitude encoding threshold. The system combines computational power constraints with geospatial constraints, and simultaneously uses the two sets of initial screening results to perform unified correction processing on the initial coverage identification results. For entries in the original coverage identification results, it simultaneously verifies whether they pass the computational complexity screening and the geospatial latitude range screening, eliminates false matching objects that do not meet either condition, and retains valid identification entries that simultaneously meet the computational power threshold and spatial range requirements. This completes the elimination, correction, and organization of the original identification results, resulting in adjusted identification results. The combined computational power overhead and geospatial constraints eliminate the mismatch problem caused by the initial quadtree retrieval, improve the accuracy of spatial identification, and provide reliable spatial analysis input for subsequent fence polygon parsing and running task generation. In this embodiment, spatial dimension filtering of latitude and longitude codes is performed by combining a second initial screening condition with a preset latitude coding threshold to eliminate abnormal geographical points that deviate from the effective monitoring area, thereby achieving refined purification of spatial data. Simultaneously, the first initial screening result based on computational complexity is integrated to correct the original coverage identification results from both computational load and geographic spatial dimensions, effectively eliminating false coverage matching entries generated during the initial spatial retrieval of the quadtree. This solution filters invalid analysis objects without reducing spatial detection coverage, reducing system geometric computation overhead and improving spatial analysis efficiency in scenarios with multiple concurrent management objects. Through dual verification and optimization of computational and spatial constraints, the accuracy and reliability of virtual fence coverage identification are improved, providing stable spatial data support for subsequent fence parsing, rule configuration, and running test instruction generation, thus enhancing the overall stability and accuracy of the campus running monitoring system.

[0022] Please see Figure 6 The sixth embodiment of a running monitoring method in this invention includes: 601. Analyze each fence polygon according to the preset fence extraction conditions, adjust the recognition results and the preset deconstruction model algorithm to obtain a set of non-covered fences; In this embodiment, the adjustment and identification results output by the previous process are used as the basis for spatial determination. Combined with preset fence extraction conditions and a deconstruction model algorithm, fence polygon filtering and analysis are completed. The fence extraction conditions define the spatial boundary threshold, polygon size constraints, and geographical range filtering rules for effective motion fences. The deconstruction model algorithm performs spatial decomposition, regional attribute identification, and intersection relationship judgment on each fence polygon. The system traverses all fence polygons, filters and removes fence areas that match the spatial coverage of the configuration management object, separates the geographical ranges that do not meet the spatial coverage relationship, and summarizes them into a set of non-covered fences. This achieves refined classification of spatial fences, distinguishing between allocable fences and overlapping fences, providing an independent fence data source for subsequent differentiated running rule configuration, and avoiding spatial conflicts between multiple object fences. 602. Configure the non-coverage fence set according to the preset running rule package and each configuration management object to obtain the configuration fence set; In this embodiment, the virtual fence not only limits the geographical area where students run, but also carries the corresponding running path management mode. The management backend pre-binds path control logic to various fences, integrating and encapsulating fence boundary range, path type, open time period, and anti-cheating strategy into a complete running rule package. For each fence within the non-covered fence set, the system completes mode matching based on the campus level to which the configured management object belongs, supporting four typical operating modes: no-point running mode, where students only need to complete mileage, pace, and time indicators within the fence, without needing to check in at a checkpoint, suitable for free exercise scenarios such as the track and field and the lakeside road; random checkpoint running mode, where the system presets checkpoints within the fence. The system employs several mechanisms: 1) Randomly selects points to generate a single task when the run starts, suppressing cheating on fixed routes; 2) Sequential route point running mode, where administrators preset the route and point order to guide students through designated safe areas and tracks; and 3) Sequential point-by-point running mode, where the next target point is unlocked only after the current point is completed, preventing students from obtaining the complete route in advance and reducing the risk of cheating through proxy running, trajectory simulation, and alternative transportation. 4) After binding the path mode, point set, point release rules, opening time, and anti-cheating level to the non-covered fence set, a configured fence set is generated, achieving the foundation for intelligent campus running management: "fences are rules, fences define paths, and path linkage prevents cheating." 603. Divide the virtual fence management area according to the set of coordinate points to obtain multiple management areas; In this embodiment, relying on the geographic point distribution information carried by the coordinate point set collected in the previous step, the virtual fence management area is spatially segmented. According to the campus tree-like hierarchical boundary, road geographic dividing line, and the natural outline of the fence polygon, the area segmentation benchmark is set, and the large-scale contiguous virtual fence management area is discretized into several spatially independent and clearly defined sub-management areas. Each sub-management area has an independent geographic identifier and jurisdiction attribute, and can be bound to configuration management objects of different grades and classes. Through area division, the large-scale campus geographic space is divided into segments for control, reducing the scope of single spatial retrieval calculation, reducing the computational overhead of subsequent fence and positioning coordinate intersection detection, and supporting the efficient operation of multi-student concurrent running monitoring scenarios. 604. Generate running test instructions based on the coverage fence set and each management area; In this embodiment, the virtual fence management area serves as the spatial carrier for the campus running task. The configured set of coverage fences is spatially associated and bound to each segmented management area. The system matches the subdivided management area to which the configured management object belongs, retrieves all rule parameters corresponding to the area, including fence boundaries, path patterns, check-in points, time constraints, and anti-cheating levels, and encapsulates them into standardized running test instructions, which are then sent to the student terminal. The running test instructions fully encompass the effective exercise range, path execution requirements, assessment indicators, and verification strategies. Upon receiving the instructions, the student terminal automatically initializes the running monitoring task, simultaneously initiating location data acquisition, motion sensor sampling, and real-time fence boundary determination. This enables the distribution of spatial areas, virtual fences, and running rules, ensuring consistency between the student terminal and the backend monitoring rules, and supporting the orderly implementation of subsequent full-process running monitoring, boundary violation identification, and layered anti-cheating verification. In this embodiment, by relying on the deconstruction model algorithm and fence extraction conditions to filter out non-covered fence sets, spatial conflicts between fences of multiple managed objects are effectively avoided, and fence resources are allocated in a refined manner. At the same time, a multi-mode running rule package is provided to adapt to diverse campus running scenarios such as free exercise, random locations, sequential check-in, and point-by-point unlocking, specifically addressing the risk of cheating on fixed routes. By splitting and subdividing the management area by coordinate point set, the spatial retrieval calculation range is reduced, improving the operational efficiency of concurrent monitoring of multiple students. Finally, by relying on the linkage between the fence set and the virtual fence management area, running test instructions are generated, realizing the unified distribution and implementation of fence boundaries, running rules, and anti-cheating strategies. This constructs an intelligent, adaptable, and highly reliable campus running fence management system, effectively improving the standardization and scenario adaptability of running monitoring.

[0023] Please see Figure 7 The seventh embodiment of a running monitoring method in this invention includes: 701. Based on the cheating risk assessment method and the set of coverage fences, evaluate and analyze the location data and motion sensor data to obtain the risk assessment results; In this embodiment, the effective movement boundary range is defined based on the coverage fence set output by the preprocessor. Multi-source fusion risk prediction is performed by fusing location data and motion sensor data to achieve real-time location monitoring and fence boundary linkage control. During the running process, the student terminal continuously outputs GPS positioning data and motion sensor data composed of accelerometers and gyroscopes at a preset acquisition frequency (e.g., once per second). The system uses the ray method or the number of turns method to determine in real time whether the current student's location coordinates fall within the virtual fence polygon corresponding to the coverage fence set, completing the location and fence space matching verification. If the current position is determined to be outside the fence boundary, the system immediately executes the running record pause mechanism, stopping the accumulation of effective running mileage, exercise duration, and real-time pace, and simultaneously displays a pop-up on the student terminal. The system displays a message stating "Exceeded the movement range"; once the location coordinates return to the interior of the fence polygon, the system automatically resumes running data collection and statistics; for interrupted periods outside the fence, the system marks the track segment as an interruption segment, which is not included in the final valid running score, but the track points are fully retained to maintain track continuity for post-event auditing and tracing; the system synchronously records the timestamp, location coordinates, and duration of each boundary crossing event, using all boundary crossing event information as a spatial dimension auxiliary feature for cheating risk assessment; combining fence entry and exit records, track point distribution, location continuity, and raw time-series data from motion sensors, the system conducts a quantitative assessment according to a preset cheating risk assessment method, outputs risk assessment results, and provides basic risk input for subsequent hierarchical and refined anti-cheating verification; 702. Verify the risk assessment results based on the multi-level anti-cheating analysis mechanism, preset time intervals, and preset distance intervals to obtain the verification analysis results; In this embodiment, a segmented progressive multi-level anti-cheating analysis mechanism is adopted to divide the complete running link into several continuous independent movement segments, and dynamically verify the authenticity of the movement based on the movement characteristics within the segment. The system completes the segmentation according to a preset time interval or preset distance interval, which can be set to generate an independent movement segment every 20 seconds, every 100 meters, or every 200 meters, for example. For each movement segment, the system extracts multi-dimensional information such as GPS position sequence, instantaneous speed, acceleration, gyroscope attitude data, steps, and cadence within the segment, calculates key movement characteristics such as average pace, average cadence, estimated stride, trajectory curvature, and displacement continuity of the segment, and implements progressive verification according to three detection levels: low, medium, and high. In the low-level detection mode, the system performs basic rule verification to check whether the total distance of a single run meets the standard, whether the pace of the whole run falls within the reasonable range preset by the school, and whether the required check-in points are completed. This is suitable for application scenarios with low intensity of daily exercise and management. In the medium-level detection mode, the consistency verification of stride length and cadence in segments is superimposed. Under real running behavior, students' cadence, stride length, and pace have a stable physiological correlation. When a segment shows abnormalities such as a pace that is too high but a cadence that is too low, a valid cadence that exists but GPS does not show a valid displacement, or a sudden and large fluctuation in stride length, the segment is marked as a suspected abnormal segment. If multiple segments are abnormal or the proportion of abnormal segments exceeds a preset threshold, an alarm is triggered and the running record is included in the manual review queue. In the high-level detection mode, GPS trajectory comprehensive feature analysis is further introduced to identify phenomena such as sudden changes in trajectory curvature, jumps in positioning points, abnormal short-distance straight-line displacement, sudden speed changes, and mismatches in the duration of point movement. This is used to identify typical non-running cheating behaviors such as cycling, riding in vehicles, virtual positioning, and trajectory simulation. Typical judgment scenarios: if GPS generates a long-distance displacement in a short period of time, but the step frequency and stride data cannot support this displacement, or if the speed continuously exceeds the human running limit and the movement posture characteristics do not match, then the segment is judged to have a high risk of cheating. The aforementioned multi-level anti-cheating mechanism supports dynamic switching configuration based on campus management needs, fence type, running route mode, and running time, which can flexibly balance the consumption of computing resources, student user experience, and anti-cheating control intensity. The system combines the segment verification conclusions to perform a second review and correction on the risk assessment results obtained in the previous stage, and finally outputs the verification analysis results, providing a layered and traceable judgment basis for generating the final running monitoring results. In this embodiment, a hierarchical and progressive campus running anti-cheating monitoring system is constructed by combining virtual fence spatial constraints with multi-source motion data fusion verification. This solution relies on a set of coverage fences to achieve linked monitoring of GPS location data and motion sensor data, real-time identification of user movement boundaries, automatic control of the start and stop of running data and effective statistics, and differentiation between compliant running trajectories and out-of-bounds interrupted trajectories, ensuring the compliance of the running space. Simultaneously, based on a time and distance dual-dimensional segmented verification mode, coupled with a low, medium, and high-level progressive anti-cheating mechanism, multi-layered verification is performed from basic exercise indicators, stride frequency and stride physiological correlation to GPS trajectory spatiotemporal characteristics, which can identify various violations such as virtual positioning, riding in vehicles, and running on behalf of others. Furthermore, this mechanism can flexibly adapt to different campus management scenarios, dynamically balance computing power consumption and anti-cheating intensity, and all abnormal data is fully traceable, improving the objectivity, accuracy, and fairness of campus running monitoring, standardizing student running behavior, and strengthening the standardization and security of campus sports assessment management.

[0024] The above describes a running monitoring method according to an embodiment of the present invention. The following describes a running monitoring device according to an embodiment of the present invention. Please refer to [link / reference]. Figure 8 One embodiment of the running monitoring device of the present invention includes: The first data acquisition module 1 is used to acquire the virtual fence configuration parameters, multiple managed objects, and the virtual fence management area of ​​the virtual fence system. Configuration module 2 is used to configure each management object according to the preset campus tree hierarchy and virtual fence configuration parameters to obtain multiple configuration management objects; Test module 3 is used to generate running test instructions based on each configuration management object and virtual fence management area, and enter the running test state according to the running test instructions; The second data acquisition module 4 is used to acquire the location data and motion sensor data of each configuration management object during the running test. Analysis module 5 is used to analyze the positioning data and motion sensor data according to the preset cheating risk assessment method and the preset multi-level anti-cheating analysis mechanism to obtain the verification analysis results. Result generation module 6 is used to generate running monitoring results based on location data, motion sensor data and verification analysis results; In this embodiment, a comprehensive campus running monitoring system was constructed, encompassing parameter configuration, command issuance, data collection, hierarchical verification, and result output. Utilizing a hierarchical campus tree structure combined with virtual fence parameters, the system enables configuration of various managed objects, adapting to diverse campus running management scenarios. Running test commands unify front-end and back-end monitoring rules, ensuring consistency between terminal data collection and back-end judgment logic. Simultaneously, GPS location data and multi-source motion sensor time-series data are integrated to fully reconstruct the spatiotemporal movement state of students during running. Coupled with a progressive, multi-level anti-cheating verification mechanism, various cheating behaviors are identified from multiple dimensions, including basic indicators, physiological correlations, and trajectory consistency, with hierarchical marking of abnormal movement risks. This entire solution achieves refined, differentiated, and intelligent management of campus running, improving the accuracy of running authenticity verification and ensuring the fairness, standardization, and credibility of campus physical education assessments.

[0025] Figure 9This is a schematic diagram of the structure of a running monitoring device 900 provided in an embodiment of the present invention. The running monitoring device 900 can vary significantly due to different configurations or performance. It may include one or more central processing units (CPUs) 910 (e.g., one or more processors) and a memory 920, and one or more storage media 930 (e.g., one or more mass storage devices) for storing application programs 933 or data 932. The memory 920 and storage media 930 can be temporary or persistent storage. The program stored in the storage media 930 may include one or more modules (not shown in the diagram), each module including a series of instruction operations on the running monitoring device 900. Furthermore, the processor 910 may be configured to communicate with the storage media 930 and execute a series of instruction operations in the storage media 930 on the running monitoring device 900 to implement the steps of the running monitoring method provided in the above-described method embodiments.

[0026] A running monitoring device 900 may further include one or more power supplies 940, one or more wired or wireless network interfaces 950, one or more input / output interfaces 960, and / or one or more operating systems 931, such as Windows Server, MacOSX, Unix, Linux, FreeBSD, etc. Those skilled in the art will understand that... Figure 9 The structure of the running monitoring device shown does not constitute a limitation on the running monitoring device, and may include more or fewer components than shown, or combine certain components, or have different component arrangements.

[0027] The present invention also provides a computer-readable storage medium, which can be a non-volatile computer-readable storage medium or a volatile computer-readable storage medium, wherein the computer-readable storage medium stores instructions that, when executed on a computer, cause the computer to perform the steps of a running monitoring method.

[0028] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the system, device, or unit described above can be referred to the corresponding process in the foregoing method embodiments, and will not be repeated here.

[0029] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0030] Finally, it should be noted that the above descriptions are merely preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. 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. A running monitoring method, characterized in that, include: Obtain the virtual fence configuration parameters, multiple managed objects, and the virtual fence management area of ​​the virtual fence system; Configure each management object according to the preset campus tree hierarchy and virtual fence configuration parameters to obtain multiple configuration management objects; The system generates running test commands based on each configuration management object and virtual fence management area, and enters the running test state based on the running test commands. During the running test, the location data and motion sensor data of each configuration management object are obtained; The positioning data and motion sensor data are analyzed based on the preset cheating risk assessment method and the preset multi-level anti-cheating analysis mechanism to obtain the verification analysis results. Running monitoring results are generated based on location data, motion sensor data, and verification analysis results.

2. The running monitoring method as described in claim 1, characterized in that, The process of generating running test instructions based on each configuration management object and virtual fence management area includes: The configuration management objects are identified and analyzed according to the preset quadtree space retrieval method to obtain the coverage identification results and multiple computational complexities; The coverage recognition results are adjusted based on the preset spatial hash collision coarse screening method, various computational complexities, preset first initial screening conditions, and preset second initial screening conditions to obtain the adjusted recognition results. Based on the adjusted recognition results and the virtual fence management area, a running test instruction is generated.

3. The running monitoring method as described in claim 2, characterized in that, The step involves identifying and analyzing each configuration management object using a preset quadtree space retrieval method to obtain coverage identification results and multiple computational complexities, including: Obtain the geographic coordinates of each configuration management object to obtain a set of coordinate points; The coordinate point set is geometrically constructed according to the preset polygon intersection detection method to obtain multiple fence polygons; The quadtree spatial retrieval method and the preset geometric length threshold are used to identify and analyze each fence polygon to obtain the coverage identification results and multiple computational complexities.

4. The running monitoring method as described in claim 3, characterized in that, The adjustment of the coverage recognition results based on the preset spatial hash collision coarse screening method, various computational complexities, preset first initial screening conditions, and preset second initial screening conditions to obtain the adjusted recognition results includes: The coordinate point set is encoded using a spatial hash collision coarse screening method and a preset global mapping dictionary to obtain multiple latitude and longitude codes; The computational complexity is screened according to the first initial screening criteria and the preset complexity threshold to obtain the first initial screening result; The coverage identification results are adjusted based on the second initial screening conditions, the first initial screening results, and the latitude and longitude codes to obtain the adjusted identification results.

5. A running monitoring method as described in claim 4, characterized in that, The adjustment of the coverage identification results based on the second initial screening conditions, the first initial screening results, and each latitude and longitude code to obtain the adjusted identification results includes: The latitude and longitude codes are filtered according to the second initial screening conditions and the preset latitude coding threshold to obtain the second initial screening results; The coverage identification results are adjusted based on the first and second preliminary screening results to obtain the adjusted identification results.

6. The running monitoring method as described in claim 3, characterized in that, The step of generating running test instructions based on the adjusted recognition results and the virtual fence management area includes: Based on the preset fence extraction conditions, the recognition results are adjusted and the preset deconstruction model algorithm is used to analyze each fence polygon to obtain a set of non-covered fences. The non-coverage fence set is pattern-configured according to the preset running rule package and each configuration management object to obtain the configuration fence set; The virtual fence management area is divided based on the set of coordinate points to obtain multiple management areas; Run test instructions are generated based on the coverage fence set and each management area.

7. A running monitoring method as described in claim 6, characterized in that, The step involves analyzing positioning data and motion sensor data based on a preset cheating risk assessment method and a preset multi-level anti-cheating analysis mechanism to obtain verification analysis results, including: The location data and motion sensor data are evaluated and analyzed based on the cheating risk assessment method and the set of coverage fences to obtain the risk assessment results; The risk assessment results are verified based on a multi-level anti-cheating analysis mechanism, preset time intervals, and preset distance intervals to obtain the verification analysis results.

8. A running monitoring device, characterized in that, include: The first data acquisition module is used to acquire the virtual fence configuration parameters, multiple managed objects, and the virtual fence management area of ​​the virtual fence system. The configuration module is used to configure each management object according to the preset campus tree hierarchy and virtual fence configuration parameters to obtain multiple configuration management objects; The testing module is used to generate running test instructions based on each configuration management object and virtual fence management area, and enter the running test state according to the running test instructions; The second data acquisition module is used to acquire the location data and motion sensor data of each configuration management object during the running test. The analysis module is used to analyze the location data and motion sensor data according to the preset cheating risk assessment method and the preset multi-level anti-cheating analysis mechanism to obtain the verification analysis results. The results generation module is used to generate running monitoring results based on location data, motion sensor data, and verification analysis results.

9. A running monitoring device, characterized in that, The running monitoring device includes: a memory and at least one processor, wherein the memory stores instructions; At least one of the processors invokes the instructions in the memory to cause the running monitoring device to perform the steps of the running monitoring method as described in any one of claims 1-7.

10. A computer-readable storage medium storing instructions thereon, characterized in that, When the instructions are executed by the processor, they implement the steps of the running monitoring method as described in any one of claims 1-7.

Images